Assessment of the variation of skin barrier function with anatomic site, age, gender and ethnicity


Majella E. Lane, Department of Pharmaceutics, The School of Pharmacy, University of London, 29-39 Brunswick Square, London, U.K. Tel.: +44 207 753 5821; fax: +44 870 1659275; e-mail:


The skin is the largest organ of the human body and its functions include protection, thermoregulation sensation and secretion. Significant advances in our understanding of how the morphology and physiology of the skin contribute to the skin’s barrier role have been achieved in recent years. The aim of this review is to summarize the principal approaches which have been used to assess variation in skin barrier function with anatomic site, age, gender, and ethnicity. The methods discussed include trans-epidermal water loss (TEWL) measurement, assessment of corneocyte size, response to vasoactive compounds and attenuated total reflectance Fourier transform infrared (ATR-FTIR) interrogation of skin. The utility of the various methods is considered and the most important findings in the literature to date are highlighted.


La peau est l’organe le plus vaste du corps humain et ses fonctions incluent protection, thermo-régulation, sens du toucher et sécrétions diverses. Récemment, d’importants progrès ont été réalisés concernant la compréhension du rôle de la morphologie et de la physiologie de la peau dans sa fonction barrière. L’objectif de cet article est de résumer les principales approches qui ont été utilisées afin d’évaluer les variations de la fonction barrière de la peau selon la site anatomique, l’âge, le genre et l’origine ethnique. Les méthodes utilisées sont les suivantes: mesure de la perte d’eau trans-épidermique [TEWL], évaluation de la taille des cornéocytes, réaction aux rubéfiants, et examen de la peau avec la spectroscopie infra-rouge à transformée de Fourier: “Attenuated Total Reflectance Fourier Transform Infrared” [ATR-FTIR]. L’utilité des différentes méthodes est examinée et les avancées les plus importantes présentées dans la littérature jusqu’à aujourd’hui sont soulignées.


The skin is the largest organ of the human body, covering an area of approximately 2 m2. It is the interface with the external environment, and prevents the penetration of foreign molecules and the loss of water and endogenous substances [1]. It is essentially composed of two major layers: the epidermis, an unvascularized layer, and the underlying dermis which contains a rich supply of capillaries, nerves, sweat and sebaceous glands and hair follicles, supported by connective tissue [2, 3]. The barrier function of the skin resides primarily in the epidermis and is localized in the stratum corneum (SC) typically 10- to 20-μm thick [4].

Inter- and intra-variability in human skin barrier as a function of anatomic site, age, gender and ethnic differences have been reported and reviewed by a number of authors [5–8]. The methods typically used to assess such variation include assessment of corneocyte characteristics, trans-epidermal water loss (TEWL) measurement, skin response to vasoactive compounds and spectroscopic analysis. This article reviews the available literature in order to assess the practical applications of these techniques to characterize such variation.

Corneocyte size

The process of terminal differentiation of the epidermis (keratinization) leads to formation of the SC which is an anucleate layer [9, 10]. The SC is a biphasic structure composed of flattened, proteinaceous corneocytes, embedded in an ordered lipid matrix. The lipids form bilayers surrounding the corneocytes, analogous to a ‘brick and mortar’ model [11]. During keratinization, nuclei and organelles disappear, lipids are expelled into the intercorneocyte space, water content decreases from 70% to approximately 15% and cells become filled with keratin [10]. The keratin content forms a filamentous network providing cohesion, flexibility and elastic recovery [3, 12]. Although some corneocytes appear pentagonal or hexagonal in shape [13], most are irregular (Fig. 1).

Figure 1.

 Corneocytes from the abdomen (adapted from Ref. [20]).

Corneocytes are held together by specialized structures known as ‘corneodesmosomes’ [13] which are composed of glycoproteins of the cadherin family namely, desmoglein and desmocollin. Within the corneocytes both glycoproteins are linked to keratin filaments [14]. In normal skin, there is a progressive reduction in the number of corneodesmosomes as corneocytes migrate towards the SC surface [13]. The cleavage of peripheral corneodesmosomes at the skin surface is accomplished by secretion of an array of serine, cysteine and aspartic proteases which are secreted into the extracellular spaces of the SC during desquamation [15]. This is a steady-state process that ensures single corneocytes are released in a very orderly and imperceptible manner at the SC surface [10, 15]. A fine balance between basal cell proliferation and corneocyte desquamation maintains the skin barrier at a constant thickness (Fig. 2).

Figure 2.

 The process of desquamation (adapted from Ref. [15]).

A number of studies have characterized corneocyte size at different anatomic sites, and have further investigated how these properties vary with age, gender and ethnic group. The actual corneocyte size is a measure of the epidermal cell proliferation and cell maturity, whereas desquamation reflects corneocyte cohesiveness [16].

Anatomic site

Plewig and Marples [17] measured corneocyte size at nine different anatomic sites in four males aged from 21 to 31 years (two Black and two Caucasian subjects). Corneocytes were removed using a detergent scrub method, stained with crystal violet and basic fuchsin, and studied under light microscopy. The smallest corneocytes were found on the forehead and on the hand and the largest cells were collected from the thigh and the axilla. Corneocytes on the abdomen were also found to be larger than the lower forearm. For the range of anatomic sites studied the surface areas varied from 746 to 1222 μm2 [16, 17]. Interestingly, Roberts and Marks [18] determined the rate at which corneocytes were shed in eight volunteers, aged from 21 to 86 years. The fastest desquamation occurred in the forearm, followed by the back, and finally the abdomen, suggesting that anatomic sites with large corneocyctes are associated with a lower desquamation rate.

Rougier et al. [5] used a detergent scrub technique to collect corneocytes at different anatomic sites from a group of six to eight male volunteers, aged 20–30 years. The rank order of the corneocyte surface area was: forearm (ventral elbow) = forearm (ventral-mid) = arm (upper-outer) = abdomen > forearm (ventral-wrist) > post-auricular > forehead. In confirmation of the findings of Plewig and Marples [17] the smallest corneocytes were found on the forehead. The authors also suggested that corneocytes were smaller in regions that were more exposed to environmental factors, that is, not protected by clothing.

Pratchyapruit et al. [19] compared the corneocyte surface areas from the eyelids, cheeks and nose of 22 Japanese volunteers (7 males and 15 females, 20–40 years). Corneocytes were sized after tape-stripping, staining with haematoxylin and eosin and observed with light microscopy. Significantly larger corneocytes were collected from the eyelid, followed by the cheek and finally by the nose. For the eyelid, the mean value for corneocyte surface area was 632.3 μm2. When compared with literature data for other anatomic sites such as the abdomen and forearm, corneocytes in the eyelid appear to be relatively small which may reflect a faster turnover at this site. For example, Plewig and Marples [17] reported a corneocyte surface area of approximately 850 μm2 for the arm and the scalp, 1050 μm2 for the abdomen and 1250 μm2 for the axilla.

More recently, Machado et al. [20] measured corneocyte sizes for 90 males and females (Caucascian and Asian subjects aged 20–60 years). Six anatomic sites were examined; ventral wrist, ventral mid-forearm close to the ventral wrist – (FA1), ventral mid-forearm close to the ventral elbow (FA2), ventral elbow, forehead and abdomen. Corneocyte surface area ranged from 825 ± 76 μm2 (forehead) to 1236 ± 95 μm2 (abdomen). Corneocyte size was ranked as follows: forehead < wrist < FA1 = FA2 = elbow < abdomen. Significantly, smaller corneocytes were observed for the forehead and wrist compared with other sites and significantly larger corneocytes were observed for the abdomen relative to all other sites. These findings are in line with those of Plewig and Marples [17] and Rougier et al. [5] who reported smaller corneocytes in facial areas relative to the forearm and abdomen.


Plewig [21] analysed corneocyte size at nine anatomic sites for two black infants (one male, one female; aged 5 and 6 months) two black male children (8–11 years), four young adult males (two Black, two Caucasian; 21–31 years) and three older Caucasian males (65–85 years). For all anatomic sites, except the forehead, corneocyte size increased significantly (P < 0.05) with age. In general, cells were smaller in infants, medium sized in children, larger in adults and largest for the oldest volunteers. For example, abdominal corneocyte sizes for the different age groups were ranked as follows: infants – 850 μm2, children – 990 μm2, young adults – 1020 μm2, adults – 1400 μm2.

Leveque et al. [22] investigated corneocyte size variation for 158 volunteers ranging in age from 8 to 89 years. A detergent scrub technique was used to collect corneocytes from the volar forearm. Corneocyte surface area increased linearly over the entire age range in line with the results obtained by Plewig [21]. The authors reported corneocyte surface area for various age groups as follows: 15 years – 800 μm2, 25 years – 900 μm2, 65 years – 1050 μm2, 75 years 1150 μm2. Analysis of the data indicates that there are significant differences between different age groups. Similar corneocyte surface areas were evident for subjects aged 35 and 65 years. However, younger subjects had significantly smaller corneocyte sizes and older subjects had significantly larger corneocyctye sizes. Rougier et al. [5] investigated corneocyte size in the upper-outer arm for three groups of six to eight male volunteers (i) 20–30 (ii) 45–55 (iii) 65–80 years. No variation in corneocyte size up to 55 years was observed. The mean corneocyte size for the 20–30 years cohort was 980 ± 34 μm2 and for the 45–55 years cohort a value of 994 ± 56 μm2 was recorded. The group aged 65–80 years did, however, show significantly larger corneocytes (1141 ± 63 μm2) relative to the other groups. A relatively small numbers of subjects were used by Rougier et al. which may explain the discrepancies when compared with the data of Leveque et al. [22].

In a more recent study, Leveque et al. [23] investigated corneocyte surface area in two groups of 20 women, aged 26 ± 6 and 61 ± 5 years, using epifluorescence microscopy. The mid-volar forearm, mid outer arm and calf were selected for analysis The forearm corneocyte size was observed to vary with the age of the volunteers with older women having significantly larger corneocytes than younger women. Although similar trends were observed for the outer arm the values were not significantly different which may reflect the low number of measurements recorded.


The variation of corneocyte size with gender was investigated by Plewig [21] for four males and three females, all aged 21–31 years. For all anatomic sites, larger corneocyte sizes were observed for women than men. Corneocyte surface areas varied from 746 to 1222 μm2 for males, and from 896 to 1346 μm2 for females. This contrasts with the data of Rougier et al. [5] who reported no gender differences in corneocyte surface area for the forehead and for the upper-outer arm for a slightly larger study group (eight males and seven females).

Fluhr et al. [24] investigated corneocyte size in 21 post-menopausal and 33 pre-menopausal women and 25 males. Corneocytes were collected from the ventral forearm using a detergent scrub technique and imaged with videomicroscopy. Significantly, smaller corneocytes were found in pre-menopausal women, compared with the other groups and there were no differences in corneocyte sizes for post-menopausal females and for males. The authors suggested that in women desquamation decreases as they pass from a pre- to a post-menopausal phase. Machado et al. [20] found no differences in corneocyte sizes for males and females in a study involving 34 females and 28 males, aged 20–60 years.


Corcuff et al. [25] studied corneocyte size and desquamation in Black, Caucasian and Asian subjects aged 18–25 years, however, the number of subjects and gender breakdown for each group was not disclosed. A detergent scrub technique was used to collect corneocytes from the upper-outer arm. No differences in corneocyte surface area were detected between the different groups and a mean value of 900 μm2 was determined. Machado et al. [20] reported no differences in corneocyte size when comparing 19 Caucasian and 15 Asian female subjects.

Transepidermal water loss (TEWL)

The SC receives water by diffusion mostly from the underlying tissues and also from the sweat glands with constant evaporation to the outside environment. The water that is constantly lost diffuses through the skin by a passive mode of transport, from the region of high water concentration inside the body to the low concentration at the surface. TEWL corresponds to the steady-state water vapour flux density passing through the SC to the exterior. Under these conditions, baseline TEWL can be described by Fick’s first law of diffusion [26]:



J = water vapour flux density – TEWL (kg m−2 s−1)

D = diffusion coefficient of water in the SC (m2 s−1)

Δc = positive concentration difference across the membrane (kg m−3)

Δz = membrane thickness (m)

Trans-epidermal water loss measurements are regarded as an indicator of barrier function and allow an assessment of any macroscopic changes in the SC barrier function [26]. A range of instrumentation has been developed to measure TEWL and both open and closed chamber devices are currently in use.

Anatomic site

Pinnagoda et al. [27] observed regional variations in TEWL and related this to differences in skin structure, particularly differences in the thickness of the epidermis and SC and differences in the regional distribution of the eccrine sweat glands, which are concentrated on the palms and soles, face and upper trunk. Rougier et al. [5] ranked TEWL values of different anatomic sites as follows: forearm (ventral elbow) < forearm (ventral-mid) < arm (upper-outer) < abdomen < forearm (ventral-wrist) < post-auricular < forehead. Lotte et al. [28] measured the TEWL values of eight male Caucasian volunteers (28 ± 2 years) on the arm, abdomen, post-auricular area and forehead. The lowest TEWL values were observed on the arm and on the abdomen, followed by the post-auricular area and forehead. TEWL was two to three times higher for the forehead than on the arm or abdomen. Den Arend et al. [29] measured volar forearm TEWL in 19 healthy subjects (gender and age range not disclosed), and observed higher values on the right arm than on the left arm (P < 0.05). Berardesca et al. [30] compared TEWL values between the volar and the dorsal sides of the forearm in 39 subjects from different racial groups. Within the same race, no differences were noted between volar and dorsal forearm.

In contrast to the findings of Den Arend et al. [29], Oestmann et al. [6] found no statistical differences between TEWL values for the left and right forearm in a study conducted on 21 volunteers (11 females, 10 males; 17–65 years). Conti et al. [31] compared TEWL values for different anatomic sites in 93 volunteers [35 males and 58 females; 2–92 years). TEWL values for the palm and sole were significantly higher than all other sites. The forehead, cheek and back of the hand showed similar results but were significantly higher than the remaining sites – dorsal forearm, volar forearm, ventral elbow, abdomen, upper back, lumbar region, buttock, pre-tibial area and calf, which showed similar results (P < 0.05).

Jang et al. [32] evaluated TEWL values at different anatomic sites and compared left and right forearm TEWL values in 24 volunteers (14 males, 10 females; 20–34 years). TEWL values were ranked as follows: palm > sole > wrist > back = calf = chin = mid-forearm = proximal-forearm (near ventral elbow). In line with the findings of Oestmann et al. [6], there were no significant differences in TEWL values from the right or left side of the forearm, palm and sole. Mid-forearm and proximal-forearm were significantly higher than the wrist (P < 0.05).

Schnetz et al. [33] compared TEWL values for different facial sites and mid-volar forearm in 11 volunteers (8 females, 3 males; 19–53 years). TEWL was significantly lower for the forearm when compared with the face. The highest TEWL value was detected for the cheek, followed by the chin and the forehead value was significantly lower than the cheek and chin (P < 0.05).

Chilcott and Farrar [34] measured TEWL values at five different sites on both volar forearm areas for eight males and nine females (18–28 years) all of whom were ‘right handed’. TEWL values were found to be significantly higher (P < 0.01) at both proximal and distal sites in comparison to midpoint sites. No differences between left and right forearms were evident as reported previously by Oestmann et al. [6] and Jang et al. [32].

Pratchyapruit et al. [19] studied 22 Japanese volunteers (7 males, 15 females; 20–40 years) and reported no statistical differences (P > 0.05) in TEWL measurements of the eyelid, nose and cheeks. Marrakchi and Maibach [35] recruited 20 volunteers (24–83 years) and measured TEWL values at seven different sites; forehead, nose, cheek, nasolabial area, perioral area, chin and upper eyelid, forearm and neck. Nasolabial and perioral TEWL values were significantly higher than other anatomic sites and the forearm TEWL value was significantly lower (P < 0.05) than all other measured areas. Machado et al. [20] measured TEWL for a range of anatomic sites in 90 Caucasian and Asian volunteers (males and females, 20–60 years) and ranked TEWL for the sites as follows: forehead > wrist > ventral mid-forearm close to the ventral wrist = ventral mid-forearm close to the ventral elbow = elbow = abdomen.

Anatomic sites where TEWL values are lower also correspond to those where larger corneocytes are found as observed by Rougier et al. [5] and Machado et al. [20]. Machado et al. [20] also determined a direct reciprocal relationship between TEWL and diffusional pathlength of water for six different anatomic sites with TEWL values tending to zero when corneocytes are infinitely large. The authors further noted that skin sites with smaller corneocytes have fewer cell layers, with shorter permeation pathlengths and higher TEWL values and that TEWL may be used to characterize the permeation routes for different anatomic sites.


As skin matures, it undergoes many morphologic and physiologic changes and thus TEWL variations might be expected with age. Hammarlund and Sedin [36] measured the chest, interscapular and buttock TEWL values in 32 infants after 25 and 39 weeks of gestation and observed an inverse relationship between TEWL and gestational age. TEWL was found to be 15 times higher in infants born after 25 weeks of gestation than in full-term infants. The authors suggested that the large differences in TEWL might be partly explained by the thinner epidermal layer and more numerous superficial skin vessels in pre-term babies. In a later study [37], the same authors noted that TEWL gradually decreased during the first weeks of life in pre-term infants, while, in full-term infants TEWL was almost unchanged during these first weeks. The authors suggested that the higher water loss in pre-term infants might be a result of their poorly developed SC and higher skin hydration.

Yosipovitch et al. [38] compared TEWL values in newborn infants (44 subjects; gestational age 37–42 weeks) during the first two days of life (at 5–10 h post-partum and again 24 h later) with TEWL values in adults (20 subjects, mean age 24 years). Measurements were taken for the forehead, upper back, flexor forearm, palms, abdomen, inguinal region and soles. TEWL values were significantly lower (P < 0.001) in neonates than in adults for the forehead, palms and soles, but were significantly higher (P < 0.0001) in neonates for the forearm. No differences were noticed for the abdomen and back. The authors associated the higher TEWL in the forearm with the predominant flexion pronation of the forearm in infants during the first days of life. The lower TEWL values in the neonates were also suggested to reflect adaptation to extrauterine life.

Kligman [39] measured the TEWL values of lateral leg and dorsal forearm skin in young (19–26 years, n = 8) and old (66–81 years, n = 6) volunteers and found no significant differences between the groups although slightly lower TEWL values were observed for the older group. Leveque et al. [22] reported a slight decrease in TEWL measurements of the ventral forearm in 158 volunteers aged 8–89 years, up to 60 years and a significant decrease (P < 0.05) in TEWL was observed in older volunteers.

Roskos and Guy [40] studied TEWL values in 22 males and females classified as young (19–42 years) and old (69–85 years). The volar forearm was occluded to prevent water loss and the time necessary for the recovery of the perturbed skin to return to baseline TEWL values was monitored. In contrast to the findings of Leveque et al. [22] and Kligman [39], baseline TEWL values did not change with increasing age. However, the recovery of TEWL values to baseline values was significantly slower in older skin suggesting that excessive hydration following occlusion is dissipated faster in younger subjects.

Conti et al. [31] measured TEWL in two groups of mixed gender; 63 subjects aged 12–60 years and 24 subjects over 60 years. TEWL values for the abdomen, buttocks and calf were significantly higher in the younger group. However, no statistical differences were found for the other anatomic sites. Wilhelm et al. [41] reported significantly lower TEWL values in aged volunteers (70.5 + 13.8 years) relative to those in young adults (26.7 + 2.8 years) for nine different anatomic sites.


Oestmann et al. [6] found no differences in TEWL measurements conducted on the forearms of 11 females and 10 males, aged 17–65 years. Similarly no differences in TEWL values between males and females were reported by Reed et al. [42] for measurements taken from the volar forearm of seven males and seven females, aged 22–38 years. Conti et al. [31] reported significantly higher TEWL values in males than in females for the cheek, upper back and calf for a study on 87 subjects (31 males, 56 females) but not for other anatomic sites. Chilcott and Farrar [34] reported significantly higher TEWL values (P < 0.05) for the volar forearm in eight male subjects compared to nine females. However, these subjects were aged 18–28 years whereas a wider age range and a greater number of subjects was studied by Conti et al. (12–60 years). In a later study conducted on the flexor forearm of 12 healthy volunteers (6 males, 6 females; mean age 24 years) no significant differences were found between gender [43]. Most of the published data suggest that gender does not influence TEWL but further studies with a greater number of subjects are needed to confirm whether there are differences between gender.


Berardesca and Maibach [44] measured TEWL values before and after applying 0.5% and 2.0% sodium lauryl sulphate solution to the backs of 9 Caucasian males (30.6 ± 8.8 years) and 10 Black males (29.9 ± 7.2 years). Values were recorded for three different areas: an untreated skin area, an area which was pre-occluded for 30 min and an area which had been delipidized. While higher TEWL values were observed for Black subjects for all areas, these differences were only statistically significant for the 0.5% solution applied to the pre-occluded area. The same experimental design was used by Berardesca and Maibach [45] to compare Caucasians (nine subjects, 30.6 ± 8.8 years) and Hispanics (seven subjects, 27.8 ± 4.5 years). No statistically significant differences were detected between the two races. In a further study, the same group [30] compared in vivo volar and dorsal forearm TEWL values for Black, Caucasian and Hispanic subjects and no differences were detected between the three races.

Kompaore et al. [46] analysed TEWL data collected from the forearm of seven Black, eight Caucasian and six Asian subjects (19 males and nine females; 23–32 years), before and after SC removal by tape-stripping. Baseline TEWL was significantly higher in Blacks and Asians (P < 0.01) when compared with Caucasians. However, no statistical differences were found between Blacks and Asians (P > 0.05). After tape-stripping, increased TEWL was noted in all groups, but the percentage increase was significantly higher in Asians (P < 0.05).

Sugino et al. [47] measured TEWL values in Caucasian, African-American, Hispano-American and Asian subjects. In decreasing order TEWL values were ranked as African-American > Caucasians > Hispano-Americans > Asians, however the authors did not comment on the statistical significance of the data.

Reed et al. [42] noted that baseline TEWL for the volar forearm of eight Caucasians and six Asians (seven males and seven females, aged 22–38 years) did not differ significantly. Warrier et al. [48] did not find any significant differences in forearm TEWL values for a group of 30 Black and 30 Caucasian women aged 18–45 years. However, TEWL was significantly lower in Blacks on the legs and cheeks. Forearm TEWL values were also lower in Black subjects but the difference was not statistically significant.

Berardesca et al. [49] recorded TEWL values from the mid-volar forearm of 10 Caucasian and eight Black women (mean age 42 ± 5 years) before and after tape-stripping. Each site was stripped 15 times and measurements were performed after every three tape-strips and after 48 h. There were no differences in TEWL baseline values and TEWL values after 48 h of recovery between the groups.

Yosipovitch et al. [50] compared skin barrier function between different Asian subpopulations and Caucasians. Baseline TEWL values, after tape-stripping and after a 3-h recovery period were measured in 39 volunteers (13 Chinese, 7 Malays, 10 Indians and 9 Caucasians) with a mean age of 34 ± 8 years. No statistical differences were detected in any of the measurements for the four ethnic groups.

Grimes et al. [51] reported no differences in volar forearm TEWL values in a study of African American females (18 subjects) and Caucasian females (19 subjects) aged between 35 and 65 years. Skin barrier integrity was further assessed in a subset of 8 volunteers (3 African American, 5 Caucasian) by measuring TEWL after application of a 5% aqueous solution of sodium lauryl sulphate to the volar forearm under occlusion for 6 h. TEWL was measured 30 min, 24 and 48 h after the patch removal. After patch removal an immediate increase in TEWL was observed for the Caucasian group but after 24 and 48 h TEWL values was similar in both groups.

Skin response to vasoactive compounds

After topical application nicotinic acid derivatives are known to induce vasodilatation of the peripheral blood capillaries located in the dermal papillae, adjacent to the epidermis–dermis junction [52]. This pharmacological action has been used to characterize in vivo skin barrier function by a number of research groups [53, 54]. The most commonly used derivative to date is methyl nicotinate (Fig. 3) which penetrates rapidly, and causes the appearance of redness that fades away in a few hours [55–57]. The induced erythema can be visually assessed by detecting its onset time or monitored using more objective methods such as laser Doppler velocimetry (LDV) and photoplethysmography.

Figure 3.

 Methyl nicotinate.

Anatomic site

Stoughton et al. [58] applied alcoholic solutions of different concentrations of nicotinic acid and nicotinates to the volar forearm of five volunteers, and noted that erythema always appeared first in the follicular and perifollicular areas. Cronin and Stoughton [53] used ethyl nicotinate to demonstrate greater permeability in the forehead, pre-sternal area and back relative to the limbs and noted that the arms were more reactive than the leg. Fountain et al. [59] applied methyl nicotinate solutions to different areas of the back and assessed erythema onset time visually. The response to methyl nicotinate was faster on the upper back than on the lower back; slower absorption was also associated with a more prolonged action. At higher temperatures (22 and 25°C) the onset time of erythema was faster than at 17°C which may have been associated with increased vasodilatation or increased methylnicotinate diffusion. Tur et al. [60] reported a higher and faster response to methyl nicotinate in the forehead relative to the forearm and palm. Issachar et al. [61] applied a 0.5% methyl nicotinate aqueous solution to 20 female volunteers, half with sensitive skin and half with normal skin, aged 18–31 years. The induced vasodilation occurred significantly more quickly for volunteers with sensitive skin. However, erythema developed significantly more quickly in follicular and perifollicular areas for both groups of subjects, and was confined to the area of application. The work of Albery and Hadgraft [62] and Albery et al. [63] indicated that methyl nicotinate in glycerol, water or glycerol/water mixtures and n-butyl and hexyl nicotinate in aqueous solutions are predominantly absorbed via the intercellular lipids. However, the varying responses for application at different anatomic sites also suggest that there is a contribution of appendageal transport to nicotinate penetration.

Tur et al. [64] investigated methyl nicotinate response at various forearm sites in eight male volunteers, aged 20–30 years. No differences were found in the response for the right and left forearm and lateral (thumb side) and medial positions, however, differences were found between proximal (near ventral elbow) and distal (near the ventral wrist) positions with the proximal site showing higher responses. The authors suggested that the results were consistent with the hypothesis that barrier function is constant on the forearm but that differential microvasculature sensitivity exists as a function of position.

Elsner and Maibach [65] applied a solution of methyl nicotinate to the forearm and vulvar sites in 11 pre-menopausal female volunteers and measured vasodilatation and blood flow using LDV. The induced erythema was shorter and less intense for vulvar sites when compared with the forearm. Slower percutaneous absorption does not explain these findings as the vulva has a reduced barrier function when compared with the forearm. The authors hypothesized that the lower response reflected a greater ‘wash out’ mechanism in the vulva, because of the higher vulvar basal blood flow relative to the forearm.

Jacobi et al. [66] measured the responses of seven male human volunteers (23.1 ± 2.7 years) to benzyl nicotinate when applied to the forearm, forehead and calf. Using LDV, higher and faster basal blood flow values were observed for the forehead compared to the forearm and calf. However, measurements of redness were not statistically different for the three sites.


Guy et al. [67] found no difference in methyl nicotinate response between young (20–30 years) and old (63–80 years) Caucasian subjects. Methyl nicotinate was applied to the volar forearm and blood vessel reactivity was measured by laser Doppler velocimetry and photoplethysmography.

Roskos et al. [68] evaluated the effects of methyl nicotinate on 13 Caucasian volunteers (7 males, 6 females) in order to evaluate the efficiency of the cutaneous microcirculation in young (20–34 years) and old (64–86 years) subjects. The sensitivity and efficiency of the cutaneous microcirculation was similar in both groups, indicating that microvessel reactivity is comparable in the ventral forearm of the different age groups.


Guy et al. [67] observed no differences in methyl nicotinate response on the volar forearm when comparing Black and Caucasian subjects (six subjects per group, aged 20–30 years). However, using the same methodology of LDV and similar sized groups as Guy et al. [67] Gean et al. [69] observed a significantly greater (P < 0.05) response in Black and Asian subjects compared with Caucasians for a range of different methyl nicotinate concentrations. No differences between the groups were observed when visual detection was used to evaluate the erythema response. Berardesca and Maibach investigated methyl nicotinate and hexyl nicotinate-induced vasodilation in nine White (30.6 ± 8.8 years) and 10 black (29.9 ± 7.2 years) male subjects [70]. Statistically significant differences (P = 0.04) were evident, with black subjects showing considerable lower responses than white subjects, recorded with LDV.

Kompaore and Tsuruta [71] applied four different nicotinate derivatives to the forearms of 28 male and female volunteers aged 23–46 years (8 Caucasian, 7 Black and 13 Asians). The lag time for the onset of eythema was ranked as follows: Asians < Caucasians < Blacks. Leopold and Maibach [56] investigated the penetration of lipophilic formulations of methyl nicotinate in four different ethnic groups (48 female volunteers aged 20–60 years with 12 subjects in each group). Methyl nicotine permeation increased in the following order: Blacks < Asians < Caucasians < Hispanics. Significant differences were found only between Blacks and Hispanics. The authors ascribed these differences to structural or functional differences in the SC of the two groups. However, as noted by Leopold and Lippold [72] it is also important to consider thermodynamic activity effects when examining different formulations in this manner.

Attenuated total reflectance-Fourier transform infrared spectroscopy

Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) is a technique used in conjunction with infra red spectroscopy that occurs when radiation propagates through a medium of higher refractive index and meets the interface of a medium of lower refractive index. If this medium of lower refractive index has absorption bands in the energy range of the incident radiation, then the penetration will result in energy loss due to absorbance. This phenomenon may then be amplified by successive reflection [73, 74]. ATR-FTIR is a particularly useful technique to investigate the lateral organization of the SC lipids and the phase behaviour of lipid membranes [75, 76]. Figure 4 highlights the most important peaks associated with the infrared (IR) spectrum of skin lipids.

Figure 4.

 Infrared spectrum of untreated human skin in vitro.

Both lipid conformational order and acyl chain packing may be monitored using methylene stretching frequencies. Most skin lipids have ordered orthorhombic and hexagonally packed domains. However, some short chain and conformationally disordered (liquid crystalline) lipids are also present [76]. Alkyl chain disorder results in a shift of the C–H stretching, and in particular the bands at 2850 and 2920 cm−1 associated with symmetric and asymmetric methylene group stretching, respectively. When the degree of disorder of the lipid alkyl chains increase, the C–H stretching frequency undergoes a blue shift to a higher wavenumber. As the CH asymmetric stretching absorbance originates almost entirely from the intercellular lipid, the area of this absorbance provides information on lipid content [77].

Two other characteristic skin bands are the amide I (1645 cm−1) and the amide II (1545 cm−1) associated with the C = O stretching and N–H bending of amide groups in proteins (keratins) and lipids (ceramides). The frequencies of these bands are largely dependent upon the structural configuration of the keratin. The intensity of the amide I band increases in the presence of water and the ratio of amide I/amide II band provides information on the water content of the SC [78]. Water content can also be evaluated by the broad band located around 3300 cm−1 that contains OH stretches of water (3200 cm−1), lipid polar heads (3400 cm−1) and NH stretches (3300 cm−1) associated with keratinized protein [79].

The 1740 cm−1 band corresponds to the C = O stretching band of lipid ester carbonyl and is mainly indicative of the presence of sebum in and on the SC. Another small shoulder can sometimes be detected at 1710 cm−1 which is associated with the C = O stretching of carboxylic acids [80].

Anatomic site

Bommannan et al. [81] studied the IR properties of human ventral forearm as a function of depth using tape-stripping. The degree of disorder of the SC intercellular lipids was found to decrease over the outer cell layers (up to three tape strips) and then to remain essentially constant. The lipid content was reduced by 60% after four strips relative to the ‘no-strip’ baseline. The IR spectra also suggested that SC hydration increased from the surface towards the SC-stratum granulosum interface.

Brancaleon et al. [82] investigated the hydration, lipid composition and conformation of the aliphatic chains of both the superficial and deeper layers of the SC. Measurements were performed on the forehead, nose, neck, fingertips and volar forearm of 18 volunteers (10 males, 8 women; 25–60 years). The intensity of bands at 3300, 2920 and 2850 cm−1 varied as a function of the anatomic site. The ratio between the intensity of the band at 2920 and 3300 cm−1 (I2920/I3300) was highest for the forehead and lowest for the forearm, indicating that the forearm had the lowest lipid content of the sites studied. The methylene group stretching frequency was also lowest for the forearm suggesting that more ordered lipids can be found in the forearm. For regions with lower sebum quantities (forearm and finger) the peaks at 1710 and 1740 cm−1 were barely detectable whereas other sites showed a small shoulder. No shifts were detected in amide bands but their intensity varied in the same way as the 3300 cm−1 band.


Puttnam [73] compared ATR-FTIR spectra collected from the palms of five adults (30–36 years) and two children (6 and 12 years). Considerable variations were observed in spectra from adults compared with children and between the ratios of the absorbances at 1640, 1550 and 1520 cm−1 bands relative to absorbance at 1540 cm−1. Gloor et al. [78] used ATR-FTIR to investigate the variation of skin hydration with age. The flexor forearm was evaluated in 31 females and 33 males aged between 12 and 69 years were studied. The ratio of amide I to amide II band was used as a measure of water content. No clear dependence on age was evident and there was greater variation in measurements from subjects over 45 and under 15 years.


Using the amide I/amide II band ratio as a measure of hydration Gloor et al. [78] found no significant differences in water content of males and females using the amide I/amide II ratio as an index of SC hydration. For a subject group of 10 males and 8 females Brancaleon et al. [82] found lower sebum levels in women. The 1710 cm−1 band was generally smaller in female subjects and the band at 1740 cm−1 was not detectable in the forearm and finger of females, whereas in men it was always detectable. The authors suggested that differences in hormonal or dietary habits were responsible for these variations.


The actual dimensions of corneocytes are clearly dependent on anatomic sites with areas such as the face having smaller corneocytes than the forearm. Variation of corneocyte size with gender is more difficult to characterize although a number of studies indicate that females tend to have larger corneocytes than males. Corneocytes are also larger in older subjects. No variation in corneocyte size is apparent in different ethnic groups. Values for TEWL appear to be generally higher for the face relative to forearm and abdomen but the palm and sole have the highest values. TEWL is inversely correlated with effective path length for water diffusion at different anatomic sites. Although a number of studies suggest that TEWL decreases with age there is no clear consensus from the published data that this is the case. The available literature suggests that TEWL values do not differ significantly between males and females or for different ethnic groups but most of the studies employ rather small sample sizes and therefore caution is needed when interpreting the findings. The response to nicotinates is generally higher and faster in facial areas relative to other anatomic sites in line with the corneocyte size and TEWL data. However, the data for nicotinate permeation in different age and ethnic groups appears contradictory. Few in vivo studies have been conducted with ATR-FTIR but variation with anatomic site, gender and age are evident. Further studies with larger numbers of volunteers are necessary to elucidate the true value of ATR-FTIR for such investigations.


We thank Fundação para a Ciência e a Tecnologia (FCT), Portugal, for funding support of this work.