Thickness, medullation and growth rate of female scalp hair are subject to significant variation according to pigmentation and

ARTICLE

Auteur(s) : Dominique VAN NESTE

Skinterface 9, rue du Sondart B-7500 Tournai, Belgium

Article accepted on 20/10/03

The biological value of human hair colour is poorly understood. Clinical experience tells us that scalp hair is an excellent sun-lock as one of the first signs of unnoticed hair loss may be an unexpected sunburn of the scalp skin. Also greying is obviously associated with ageing. Other modifications of scalp hair associated with ageing are usually described as hair “thinning”. Because this is a qualitative term describing the global perception of scalp hair it cannot help our understanding of the basic modification of hair growth during ageing. The specificity of “senescent alopecia” remains a matter of dispute [1, 2]. A detailed knowledge of scalp behaviour during ageing requires measurement of four variables known to quantitatively describe the hairiness of scalp namely, number of hairs (and hair follicles) per unit scalp area, growth phase duration, growth rate and thickness [3]. The two last variables can be measured using light microscopy irrespective of hair pigmentation and a correlation was shown between these variables in various conditions [4, 5].
Little is known about the relationship between hair pigmentation, growth rate and thickness and the possible changes occurring with age. We have now characterised these hair growth variables in relation to hair pigmentation in postmenopausal females and compared the results with those from young and mature female controls.

Materials and methods

Volunteers

24 women agreed to participate in a hair physiology study after approval of the protocol by the Ethical Committee. 8 subjects (average age 60.5 years; range 50-65) from a group of 12 untreated post menopausal Caucasian females with “thinning” hair showed both pigmented (P) and unpigmented (white-W) hairs on the same scalp. Hair loss was classified according to Ludwig's classification [6].
7 young (average age 19 years; range 15-21) and 5 mature (average age 42.8 years; range 38-48) healthy females served as controls. Their hair colour ranged from light to dark brown with few if any white hairs.

Scalp Sites

All hairs from 3 1.2 × 1.0 cm scalp sites, 2 on the top of the head (T), one occipital (O) in each individual were clipped close to the scalp surface. The patients returned one month later (average 29.2 days; range 28-33 days) for clipping and collection of the grown hairs. Menopausal women were sampled twice at a one month interval. Clipped hairs were mounted with double sided adhesive tape on microscope slides and length and thickness (average of maximum and minimum diameter; precision of the objective ruler: 0.1 µm) were determined microscopically. The presence of medulla was evaluated against a standardized ruler and medulla was described as occupying 10, 20, 50, 70 or 90% of the length of the hair segment. Only those hairs with two clear cut ends at both sites were measured (precision of length measures 0.5 mm). W hairs were defined as indistinguishable in colour from background when placed against a white card (Fig. 1). Other hairs were considered P.

Statistical Analysis

After checking of our samples for normality of data distribution, analysis of variance was used for the comparison of W and P hairs and of P hairs from different subject groups. Comparison of P and W hairs of similar thickness was made by selecting hairs to form pairs of P and W hairs each from the same subject within 10µm bands of thickness. Growth rates were compared for each thickness band by a paired t-test. p values < 0.05 were considered as statistically significant. Study of pigmented hairs only was performed in different age groups using ANOVA and the descriptive statistics (average and standard deviation) are given for each variable in a table.

Results

Comparison of measurements in postmenopausal women with pigmented and unpigmented hair

Clinical demographics are shown in Table I. There was no correlation between age, Ludwig category and distribution of W hairs in the postmenopausal females. W hairs were expressed more on T sites.

This table lists severity of hair loss (Ludwig scale), age and proportion of white hair (% W hairs) in postmenopausal women.
Although no difference (p > 0.05) was observed between sites for all hairs (Fig. 2), the average thickness of W hair for all sites significantly (p < 0.0001) exceeded that of P hair. Both T and O sites showed similar differences.
As shown in figure 3, the medulla of W hair for all sites was significantly (p < 0.0001) more developed than the medulla of P hair. Medullas in T sites were more developed than O sites (p < 0.05) but this was restricted to W hair.
Growth rate of W hair was significantly higher than that of P hair (p < 0.0001; Fig. 4). There was also a significant variation with scalp site (p < 0.0001). Interaction between site and colour was significant (p = 0.0062) with the following rank order O/W > T/W > O/P > T/P.
As thickness is known to influence growth rate independently of pigmentation, we analysed in greater detail the effect of hair pigmentation on hair growth rate for hairs of equal thickness. Comparison of pairs of P and W hairs of the same thickness range showed that W hairs grew significantly faster (average 10% difference in each subgroup; p < 0.001) than P hairs in the main terminal thickness range of 50-70 µm (90 paired observations; 34 in the 50-60 µm, 30 in the 60-70 µm and 26 in the 70-80 µm). P and W hairs displayed similar rates of growth in the thickest hair (> 80 µm) and thinner hair (< 50 µm) categories. Thus faster rate in growth rate of W hairs is not exclusively explained by the average increase in their thickness.

Comparison of pigmented hairs from young, mature and postmenopausal females (Fig. 5 and Table II).

Table II. Hair thickness, medulla and growth rate (mean ± SD) for Young, Mature and Postmenopausal Females

Pigmented hairs of postmenopausal females grew significantly slower than young (p < 0.05) and mature (p < 0.05) females. Hair thickness was also less than young (p < 0.05) and mature (p < 0.05) females. Medulla development was not significantly different (p > 0.05) between groups. Combined reduction of thickness and growth rate will result in a slowing down of hair production and significantly contribute to hair thinning.
White hairs from postmenopausal females had thickness and growth rates similar to control subjects but were characterised by a thicker medulla.

Discussion

The biological importance and/or significance of human hair colour is unknown even though greying is obviously associated with ageing. A rule of thumb suggests that 50% of people have 50% of grey hair by the age of 50 years [7] and premature greying has been thought to be a potential risk factor for ageing associated conditions. From a dermatological and clinical perspective, W hairs show a relative disease resistance in alopecia areata [8]. The mechanisms by which W hairs are spared these changes are not understood and may be important for the role of pigmentation and prognostic value of greying.
Hair growth rate and thickness are known to be interrelated. In male androgenetic alopecia thinner hair also grows more slowly than thicker hair [9, 10]. Female body hair after antiandrogen therapy also thins and shows reduced growth rates [4]. In menopausal females our data show that the slowing down of growth rates may also reflect clinical hair thinning. However, for hairs of equal thickness, this phenomenon also appears to affect more specifically P scalp hairs. Indeed unpigmented hair of menopausal woman maintains growth rates found in pigmented hairs in the younger age groups. To the best of our knowledge, similar studies in scalp hair of younger and ageing males have not been performed. The biology requires further investigation as to the regional variability and the potential influence of androgens or other hormonal factors responsible for unravelling these regional variations [11].
In this study as well as in a limited observation of male beard hair [12], we observed a 10% slower rate of pigmented hair growth as compared pairwise with white hair in the thickness range of 50 to 80 µm. It has been reported that W beard hair may grow up to three times faster than P hairs [5]. It is not yet clear whether this far outranging difference is due to whisker or beard hair cycle differences, to ageing or to some methodological bias. Taking our measurements into account and supposing that the thickest white hair (80 µm) was compared with the thinnest pigmented hair (50 µm) we found a 30% faster growth rate of white hair; even so this remains far lower than the findings of Nagl [5].
Our findings are in line with other authors showing that W hairs were also found thicker and displaying a more developed medulla [13]. These authors also found increased sensitivity to weathering, increased cysteic acid residues and decreased cystine and increased fibre reactivity to reducing and oxidizing agents. Whether these chemical differences in white hair, probably related in part to the lack of melanins and to the huge medulla, are also directly responsible for other white hair biophysical properties (coarseness, relative resistance to hair setting and colouring), is not clearly established.
Hair colour significantly affects visibility of hair with photographic methods [14]. This may have consequences on data generated during hair loss trials involving balding subjects as hair thinning is usually associated with less hair pigmentation [9] even though less pigmented Caucasians are usually excluded from clinical trials. Using such optical approaches in the investigation of physiological changes of hairiness or hair growth responses to drugs in ageing subjects also raises some doubts as some hair may not be detected at first sight or growth promoters might – by chance – simply affect hair pigmentation and incidently make hair more visible to the observer without modifying any of the growth variables (rate, duration of anagen, thickness, duration of the lag phase).
Nowadays a high quality contrast enhanced phototrichogram (CE-PTG) technique as shown in Figure 6 has an equal resolution power to that of the microscopical method used in the present study. Our CE-PTG techniques have now been validated for a detailed analysis of all hair variables including monitoring of all transitions of the hair growth cycle even when severe thinning and/or greying hair is present [15, 16]. Clinical trials using such refined methods would need to be considered seriously even when previous data did not show significant effects. Indeed a whole range of biological modifiers may provide investigators with clinically relevant information and possible therapeutic answers to recover age related losses of structures and functions that bother our patients suffering from hair loss. n

References

1. Kligman AM. The comparative histopathology of male-pattern baldness and senescent baldness. Clin Dermatol 1988; 6(4): 108-18.

2. Sinclair R, Jolley D, Mallari Magee J, Tosti A, Piraccini BM, Vincenzi C. Happle R, Ferrando J, Leroy T, Grimalt R, Van Neste D, Zlotogorski A, Christiano AM, Whiting D. Morphological Approach to Hair Disorders. J Invest Dermatol 2003; 8: 56-64.

3. Van Neste D. Assessment of hair loss. Clinical relevance of hair growth evaluation methods. Clin Exp Dermatol 2002; 27: 358-65.

4. Orfanos CE, Happle R. Hair and hair disorders. Germany: Springer-Verlag, 1990: 1-1057.

5. Nagl W. Different growth rates of pigmented and white hair in the beard: differentiation vs. proliferation? Br J Dermatol 1995; 132: 94-7.

6. Ludwig E. Classification of the types of androgenetic alopecia (common baldness) occuring in the female sex. Br J Dermatol 1977; 97: 247-54.

7. Keogh EV, Walsh RJ. Rate of greying of human hair. Nature 1965; 207(999): 877-8. (abstract).

8. Dawber R, Van Neste D. Hair and Scalp Disorders. Common Presenting Signs, Differential Diagnosis and Treatment. London: Martin Dunitz, 1995.

9. Runne U, Martin H. Veränderungen von telogenrate, haardichte, haardurchmesser und wachtumsgeschwindigkeit bei der androgenetischen alopezie des mannes. Der Hautarzt 1986; 37: 198-204.

10. Van Neste D, De Brouwer B, Dumortier M. Reduced linear hair growth rates of vellus and of terminal hairs produced by human balding scalp grafted onto nude mice. Ann NY Acad Sci 1991; 642: 480-2.

11. Sawaya ME, Price VH. Different levels of 5 a-Reductase Type I and II, aromatase, and androgen receptor in hair follicles of women and men with androgenetic alopecia. J Invest Dermatol 1997; 109: 296-300.

12. Van Neste D, Tobin DJ: Hair cycle and hair pigmentation: dynamic interactions and changes associated with aging. Micron, in press.

13. Hollfelder B, Blankenburg G, Wolfram LJ, Höcker H. Chemical and physical properties of pigmented and non-pigmented hair (“grey hair”). Int J Cosmet Sci 1995; 17: 87-9.

14. Rushton DH, De Brouwer B, De Coster W, Van Neste D. Comparative evaluation of scalp hair by phototrichogram and unit area trichogram analysis within the same subjects. Acta Derm Venereol (Stockh) 1993; 73: 150-3.

15. Leroy T, Van Neste D. Contrast enhanced phototrichogram pinpoints scalp hair changes in androgen sensitive areas of male androgenetic alopecia. Skin Research and Technology 2002; 8: 106-11.

16. Van Neste D, Demortier Y. Detailed monitoring of hair cycle transitions in vivo using contrast enhanced phototrichogram (CE-PTG). in Van Neste D, Blume-Peytavi U, Grimalt R, Messenger A. Hair Science and Technology. Tournai: Skinterface, 2003: pp 211-221.