Content uploaded by José Aguilera
Author content
All content in this area was uploaded by José Aguilera on Feb 19, 2024
Content may be subject to copyright.
Photochemistry and Photobiology, 2015, 91: 966–970
Human Hair as a Natural Sun Protection Agent: A Quantitative Study
Mar
ıa Victoria de G
alvez*
1
, Jos
e Aguilera
1
, Jean-Luc Bernab
o
2
, Cristina S
anchez-Rold
an
1
and
Enrique Herrera-Ceballos
1,2
1
Photobiological Dermatology Laboratory, Medical Research Centre, Department of Dermatology and Medicine,
Faculty of Medicine, University of M
alaga, Malaga, Spain
2
Universitary Hospital Virgen de la Victoria, University of M
alaga, Malaga, Spain
Received 12 November 2014, accepted 3 February 2015, DOI: 10.1111/php.12433
ABSTRACT
The rising incidence of skin cancers attributable to excessive
sun exposure has become a major health concern worldwide.
While numerous studies have analyzed the sun protective
effect of sunscreens, clothing and antioxidants, none to date
have measured the photoprotective effect of hair, despite clin-
ical evidence that individuals with balding or thinning hair
are at greater risk of skin lesions that can progress to cancer,
hence the recommendation to use hats or umbrellas. We ana-
lyzed the level of protection offered by hair according to hair
density, thickness and color using the spectral transmittance
and corrected for relative erythema effectiveness. Our results
show that hair provides a barrier against both UVB and
UVA radiation which is significantly increased with respect
to the hair density, thickness and the presence of melanins.
This is the first study to quantify sun protection factor
offered by hair, namely hair ultraviolet protection factor
(HUPF). We believe that hair should be recognized as an
important natural sun barrier in the prevention of UV-
induced skin cancers.
INTRODUCTION
The incidence of skin cancers is growing (1,2), and overexposure
to solar UV radiation is recognized as the main causative factor
(3). Lifestyle changes in recent decades have led to increased
participation in outdoor pursuits and as a consequence higher
levels of sun exposure among the general population. These
changes, combined with rising life expectancy and an aging pop-
ulation, have increased the risk of UV-induced skin cancers
(4,5).
Melanin, keratin, urocanic acid and, more recently, antioxi-
dants, are all recognized as natural photoprotective agents (6,7),
but scant attention has been paid to one of our most important
natural defenses against the sun: hair.
The scalp is one of the most heavily sun-exposed areas of the
body, and there is clinical evidence that chronic exposure to sun-
light favors the development of precancerous skin lesions and
skin cancers (3). Field cancerization of the scalp should thus be
an important consideration in skin cancers prevention and treat-
ment. While full sun protection is one of the cornerstones of
actions aimed at preventing UV-induced skin damage, and con-
siderable advances have been made in our understanding and use
of protective clothing, hats and sunscreens (8–10), the level of
protection afforded by hair has never been measured. The obser-
vation of clear differences in the incidence of head and neck
melanoma between men and women has led to the hypothesis
that hair might have a protective role in melanoma (11). It has
even been shown that sun-exposed areas of the head receive
higher levels of radiation than unexposed areas (12). Neverthe-
less, we are unaware of any studies that have analyzed, either
quantitatively or qualitatively, the photoprotective efficiency of
hair, despite the obvious value such information would have for
actions targeting the prevention of UV-induced skin cancers.
In this study, we analyze for the first time, the level of photo-
protection offered by hair of varying densities, thickness and col-
ors, as well as it has been estimated the protection factor based
on the ability to block erythema effective radiation.
METHODS
We analyzed samples of hairs of similar thickness (73–75 lm) taken
from the scalp and classified by color (brown, blond, red and white). The
samples were clean and free of any dyes or hair products, such as sham-
poo, conditioner or gel that could alter the natural transmission of UV
radiation.
The effect of hair density (as quantity of hair/volume) on photoprotec-
tive efficiency was calculated according to weight and number of hairs
by volume.
Density by weight. To analyze the level of sun protection offered by the
actual structure of hair, or simulated hair in random arrangement, samples
of healthy human brown hair were cut in 3 to 5 mm pieces in length and
randomly placed, in increasing quantities, in a cylindrical container with
a surface area of 0.785 cm
2
and a height of 1.27 cm, in order to ensure
1cm
3
. Prior to measurement of the transmission of light, each sample
was weighed on a 0.1-mg precision scale (Sartorius BL120S, Sartorius
AG G€
ottingen, Germany), and the container was shaken to ensure ran-
dom arrangement.
Density by number of hairs. To simulate the natural arrangement of
hair on the scalp (as after brushing or styling), we used samples of hair
measuring 40 mm in length taken from individuals with straight hair.
The number of hairs per sample was estimated by weight, taking as a ref-
erence the mean weight of 10 samples of 50 hairs measuring 40 mm in
length. The hairs, unstretched, were placed uniformly (to ensure a parallel
alignment) over the aperture of the light sensor (1 cm
2
) and held in place
by a plastic cuvette-like structure of 1 cm height to retain the three-
dimensional structure. The hairs were, therefore, assembled parallel to the
surface of the sphere and perpendicular to the beam of incident light. To
compare measurements of HUPF related to hair number cm
3
with
respect to hair density by weight, the weight of selected numbers of
40 mm hair was divided by 4.
Hair thickness. To analyze the effect of the hair thickness in their
photoprotective efficiency, strands of brown hair were selected from three
*Corresponding author email: mga@uma.es (Mar
ıa Victoria de G
alvez)
©2015 The American Society of Photobiology
966
different thicknesses. For each group, 5 mm small hair pieces were cut,
placed on a slide and visualized by light microscopy 40x (Nikon Eclipse
E400, Nikon Instruments Europe BV, Amsterdam, Netherlands). Micros-
copy images were taken and thickness measurement was calculated with
an image analysis program (v 6.3 Visilog Noesis, France). A total of 30
measurements of thickness in each hair group were made with mean val-
ues of 85.4 9.12 lm, 76.4 15.06 lm and 67.17 10.83 lm,
respectively.
Calculation of hair ultraviolet protection factor (HUPF). To analyze
the photoprotective efficiency of hair according to color and hair density,
we used a 300-W Oriel solar simulator (Newport Corporation, Irvine)
and a double monochromator spectroradiometer (Macam SR-9910-V7,
Irradian Co., UK) to measure the light transmitted through the hair. The
spectroradiometer was equipped with an Ulbricht sphere sensor placed
directly under the hair to ensure uniform measurement of both direct and
diffuse radiation. Both solar simulator and the Ulbrich sphere were fixed
in order to make comparable measurements.
The HUPF was estimated using the transmittance of each group of
hair at different wavelengths between 290 and 400 nm corrected for rela-
tive erythema effectiveness based on the erythema action spectrum pro-
posed by the Commission Internationale de l’Eclairage (CIE;
International Commission on Illumination) (13).
In brief, for each wavelength, the spectral irradiance was multiplied
by the relative erythema spectral effectiveness and by the hair transmit-
tance value. The HUPF was calculated using the following formula:
HUPF ¼R400
290 Ekek
R400
290 EkekTk
where HUPF is the ultraviolet protection factor of hair, E, solar spectral
irradiance, e, relative erythema spectral effectiveness (CIE) and T, spec-
tral transmittance through hair.
Statistical analysis. For each sample of hair, HUPF values were calcu-
lated from the mean SD of 5 replicas.
The HUPF values in relation to hair density were fitted to an exponential
model using Microsoft Excel 2010. Differences between regression slopes
were examined by covariance analysis. Statistical significance was set at a
Pvalue of less than 0.05 (14).
RESULTS
The estimated transmittance values for each of the UV wave-
lengths showed that the hair analyzed attenuated radiation across
the whole UV spectrum which is different depending on hair
color (Fig. 1). Light transmitted across the brown and red hair
showed a slight, but not statistically significant, increase in trans-
mittance along the whole UV spectrum with difference of trans-
mittance between 290 and 400 nm less than 5%, thus concluded
as homogeneous spectral UV transmittance. However, blond and
white hair produced a gradual, but significant elevation of trans-
mittance, especially in white hair, up to 60% at 400 nm com-
pared to 290 nm.
When hair density was increased, we observed a gradual
reduction in transmitted radiation (Fig. 2). However, the hair
density did not affect spectral transmittance with respect to dif-
ferent wavelengths as showed in the case of brown hair, with
parallel curves related to different hair densities.
Using the transmittance data, we calculated HUPF according
to density assessed by weight and by number of hairs cm
3
.In
all cases, the value increased exponentially with increasing den-
sity (Fig. 3). In order to compare the attenuation of UV radiation
of hair according to random arrangement or parallel alignment
(as hair brushing), we calculated the HUPF of brown hair in
terms of mg hairs cm
3
with respect to number of hairs cm
3
(Fig. 3A,B). Results showed higher HUPF in aligned hair com-
pared to randomly arranged hair. For example, HUPF of 500
brown aligned hairs cm
3
was 32 (according to model of
Fig. 3B), while its corresponding weight cm
3
was 22.5 mg.
According to the model of the Fig. 3A, a weight of 22.5 mg cor-
responded to a HUPF of 19.
Hair color influenced the slope values of the exponential mod-
els relating hair density assessed by number of hairs cm
3
com-
pared to HUPF, with similar slopes from brown, red and blond
hair (Fig. 3B–D), while white hair showed significantly
(P<0.05) lower slopes (Fig. 3E).
According to the hair thickness, results showed exponential
curves of HUPF significantly increased with respect to of hair
density (in terms of number of hairs cm
3
). Higher curve slope
was observed in thicker hair (85.4 lm) than lower values
observed for thinner hair (67.2 lm) (Fig. 4).
Figure 1. Spectral transmittance between 290–400 nm for different hair
colors.
Figure 2. Spectral transmittance (290–400 nm) for different brown hair
densities (measured by numbers of hair cm
3
).
Photochemistry and Photobiology, 2015, 91 967
Using the exponential fit models relating HUPF to hair den-
sity, we estimated the number of hairs by color that would be
needed per cm
3
to offer similar levels of protection to those used
for sunscreens according to the sun protection factor (SPF) cate-
gories in the European Standard on sun protection test methods
for cosmetics (15). Approximately 300 hairs cm
3
would be
needed to achieve a HUPF of 10 in the case of brown, blond
and red hair, while 400 hairs cm
3
of white hair would be
needed. To achieve a HUPF of over 40, 550 hairs cm
3
would
be required in colored hair (brown, blond and red) and approxi-
mately 700 white hairs cm
3
(Table 1).
DISCUSSION
Promoting awareness of the importance of full sun protection is
an important public health issue worldwide. Considerable
advances have been made in our understanding of artificial sun
protection products, such as fabrics, sunscreens and sun protec-
tion supplements. While it is known sun protection by hats (e.g.
wearing a 7 cm brim hat reduces the incidence solar radiation by
a factor 5 in different parts of the face and neck) (9), little is
known about the photoprotective function of human hair, even
though sun exposure is known to increase the risk of skin
Figure 3. Exponential increase in HUPF provided according to hair density and color. (A) Brown hair density assessed by weight (random
arrangement) (B–E) Hair density assessed by number of hairs cm
3
(parallel alignment). (B) Brown, (C) Blond, (D) Red, (E) White
.
968 Mar
ıa Victoria de G
alvez et al.
cancers. We have performed what is, to our knowledge, the first-
ever study to quantify, using spectrometric techniques, the ability
of the hair to protect against erythema damage induced by UV
radiation. Hair density is a determining factor in the estimation
of the UV protection offered by hair and secondly it has been
shown how hair color and thickness affect such protection.
The scalp is the highest part of the body and as such is
exposed to the greatest amount of solar radiation. Furthermore,
when a person is standing, his or her scalp is perpendicular to
the sun’s rays and therefore receives high doses of radiation,
probably explaining the relatively high incidence of precancerous
lesions and skin cancers in this area (16). The incidence of mela-
noma is rising and melanoma is less common in hair-covered
areas of the head and neck compared with bare areas, suggesting
a protective role for hair (11). In 2006, Green et al. (12) exam-
ined the hypothesis that hair offered protection against UV radia-
tion in a study in which UV exposure was analyzed using
manikin head forms equipped with UV dosimeters on the ears
and nose. Three of the manikins were fitted with human hair-
pieces and one was left bare. The findings showed that the bare
manikin received 80% more solar UV radiation than the hair-
covered manikins.
The results of our study confirm that hair is an important sun
protection agent: it reduced the transmission of UV radiation
across the spectrum, highlighting its importance as a natural pro-
tector against sun-induced erythema. This protective effect would
appear to be supported by clinical evidence, as sun-induced
lesions are much more common in bare areas of the scalp (16).
We also found that the level of protection increased exponentially
with hair density, in a manner similar to that observed for fabric
cover in a recent study by our group (17). Otherwise, it has been
observed that the hair thickness changes its photoprotection effi-
ciency which is greater in thick hair. This fact is of particular
interest in cases of androgenic alopecia where the hair miniaturi-
zation occurs over the years, and therefore the hair losses photo-
protective efficacy. Moreover, we have observed that certain
degree of order (parallel alignment) showed higher values of
HUPF compared to random arrangement of hairs. This is obvious
due that ordered alignment may confer higher cover than random
arrangement because the loss of effectiveness by overlap of hair.
On the other hand, the presence of melanin, that confers color
to the hair, adds photoprotective effect to the hair structure. We
have observed higher levels of light transmittance in white hair
as compared to colored hair (brown, red or blond). Our findings
show that the brown and red hair analyzed offered highly effec-
tive sun protection, as it absorbed similar amounts of UV radia-
tion, including UVA radiation, across the spectrum analyzed. It
is important to point out the high degree of photoprotection by
red hair (with a higher content of pheomelanin) as it is present
in skin type I in which there is a higher prevalence of skin can-
cers. However, the blond and so much more pronounced white
hair, produce a gradual increase in transmittance over the wave-
lengths of the UV spectrum, especially in the range of UVA, up
to 60% from 290 to 400 nm. These findings could have clinical
significance, as UVA radiation emits 20 times more radiant
energy than UVB radiation, and it has been widely documented
that the former has a role in actinic damage, primarily through
the generation of oxidative stress, which is associated with skin
aging and, more importantly, photocarcinogenesis (18,19).
We used the exponential fit model relating HUPF to hair den-
sity to estimate the density required to provide the level of pro-
tection similar to the categories for sunscreen products. It has to
be pointed out that solar protection factor for sunscreens is based
on in vivo end point and HUPF is based on in vitro measure-
ments considering erythema effectiveness. HUPF was related to
hair density defined as the number of hairs cm
3
in a parallel
arrangement. The number of hairs cm
2
of hair surface has been
estimated at between 0 and approximately 200 (20), meaning that
for a thickness of 1 cm, the maximum hair density would be 200
hairs cm
3
. Nonetheless, higher densities can occur in longer
hair, as parts of the scalp would be covered by the hair growing
in this area plus overlapping hair from other areas. According to
the models approximately 300 hairs cm
3
would be needed to
achieve a HUPF of 10 in the case of brown, blond and red hair,
while would be needed approximately 400 hairs cm
3
of white
hair. A HUPF value of 0–10 might appear low if we consider
the corresponding protection offered by an SPF of 0–10 in a sun-
screen. However, it should be noted that SPF ratings are calcu-
lated for an application thickness of 2 mg cm
2
, which is
unrealistic, as most people apply significantly thinner layers
(21,22). Furthermore, sunscreen products degrade with time and
can be rubbed or washed off by wear and sweat. The protection
offered by hair, however, does not vary on exposure to sun.
Figure 4. Exponential increase in HUPF provided according to three dif-
ferent brown hair thicknesses (mean values of thickness included in the
legend).
Table 1. Solar protection level related to hair density and color.
Solar Protection Level
Number of hairs cm
3
Brown Blond Red White
Low sun protection
(HUPF 0–10)
0–303 0–311 0–310 0–389
Moderate sun protection
(HUPF 10–25)
304–467 312–479 311–482 390–602
High sun protection
(HUPF 25–40)
468–551 480–565 482–570 603–711
Very high sun protection
(HUPF >40)
>551 >565 >570 >711
Photochemistry and Photobiology, 2015, 91 969
In summary, we have shown that hair is an important natural
protective agent against UVB and UVA radiation and should be
considered in the prevention of UV-induced cancer of the scalp,
particularly to consider in people with alopecia in which the use
of hats would be specially indicated for minimizer the harmful
effect of the ultraviolet radiation.
REFERENCES
1. Gray-Schopfer, V., C. Wellbrock and R. Marais (2007) Melanoma
biology and new targeted therapy. Nature 445, 851–857.
2. Aceituno-Madera, P., A. Buend
ıa-Eisman, S. Arias-Santiago and S.
Serrano-Ortega (2010) Changes in the incidence of skin cancer
between 1978 and 2002. Actas Dermosifiliogr 101,39–46.
3. International Agency for Research on Cancer (1992) IARC Mono-
graphs on the Evaluation of Carcinogenic Risks to Humans, Vol. 55.
Solar and ultraviolet radiation. International Agency for Research on
Cancer, IARC, Lyon.
4. Diaz, J. H. (2013) Updates for responsible sun exposure behavior
and photoprotection in the south. J. La State Med. Soc. 165, 277–
282.
5. Tsatsou, F., M. Trakatelli, A. Patsatsi, K. Kalokasidis and D. Sotiria-
dis (2012) Extrinsic aging: UV-mediated skin carcinogenesis. Der-
matoendocrinol. 4, 285–297.
6. Chen, L., J. Y. Hu and S. Q. Wang (2012) The role of antioxidants
in photoprotection: A critical review. J. Am. Acad. Dermatol. 67,
1013–1024.
7. De G
alvez, M. V. (2010) Antioxidants in photoprotection: Do they
really work? Actas Dermosifiliogr 101, 197–200.
8. Almutawa, F. and H. Buabbas (2014) Photoprotection: Clothing and
glass. Dermatol. Clin. 32, 439–448.
9. Diffey, B. L. and J. Cheeseman (1992) Sun protection with hats. Br.
J. Dermatol. 127,10–12.
10. Centini, M., G. Miraglia, V. Quaranta, A. Buonocore and C. Ansel-
mi (2014) Improving photoprotection: 4-methylbenzylidene camphor
microspheres. Pharm. Dev. Technol. 22,1–9.
11. Lesage, C., C. Barbe, A. Le Clainche, F. X. Lesage, P. Bernard and
F. Grange (2013) Sex-related location of head and neck melanoma
strongly argues for a major role of sun exposure in cars and photo-
protection by hair. J. Invest. Dermatol. 133, 1205–1211.
12. Green, A. C., M. Kimlin, V. Siskind and D. C. Whiteman (2006)
Hypothesis: Hair cover can protect against invasive melanoma on the
head and neck (Australia). Cancer Causes Control 17, 1263–1266.
13. ISO 17166: 1999. Erythema Reference Action Spectrum and Stan-
dard Erythema Dose. Standard CIE S 007-1998.
14. Sokal, R. R. and F. J. Rolf (1981) Biometry. Freeman, San Fran-
cisco.
15. ISO 24444: 2010. Cosmetics -Sun protection test methods- In vivo
determination of the sun protection factor (SPF). European Stan-
dardisation Organisation (CEN).
16. Youl, P. H., M. Janda, J. F. Aitken, C. B. Del Mar, D. C. Whiteman
and P. D. Baade (2011) Body-site distribution of skin cancer, pre-
malignant and common benign pigmented lesions excised in general
practice. Br. J. Dermatol. 165,35–43.
17. Aguilera, J., M. V. De G
alvez, C. S
anchez-Rold
an and E. Herrera-
Ceballos (2014) New advances in protection against solar ultraviolet
radiation in textiles for summer clothing. Photochem. Photobiol. 90,
1199–1206.
18. Gruijl, F. R. (2000) Photocarcinogenesis: UVA vs UVB. Methods
Enzymol. 319, 359–366.
19. Vile, G. F. and R. M. Tyrell (1995) UVA radiation induced oxidative
damage to lipids and proteins in vitro and in human skin fibroblasts is
dependent on iron and singlet oxygen. Radic. Biol. Med. 18, 721–730.
20. Jimenez, F. and J. M. Ruifern
andez (1999) Distribution of human
hair in follicular units. A mathematical model for estimating the
donor size in follicular unit transplantation. Dermatol. Surg. 25,
294–298.
21. Bech-Thomsen, N. and H. C. Wulf (1992) -1993) Sunbathers’appli-
cation of sunscreen is probably inadequate to obtain the sun protec-
tion factor assigned to the preparation. Photodermatol.
Photoimmunol. Photomed. 9, 242–244.
22. Neale, R., G. Williams and A. Green (2002) Application patterns
among participants randomized to daily sunscreen use in a skin can-
cer prevention trial. Arch. Dermatol. 138, 1319–1325.
970 Mar
ıa Victoria de G
alvez et al.
