British lournal of Dermatology (1991) 124. 146-1 S1
ADONIS OOO7O96391OOO33Q
Seasonal changes in human hair growth VALER[K A.RANDALL AND F.).G.EBLING* Department of Biomedical Sciences, University of Bradford *Sub-ikpartmenl of Dermatology, Academic Division of Medicine. University of Sheffield. Royal Haiiamshire Hospital U.K.
Sbeffieid.
Accepted for publication 24 August 1990
Summary
Various parameters of hair growth were determined every 28 days for 18 months in 14 healthy Caucasian men aged 18-39 with indoor occupations in Sheffield, U.K. (latitude 5 1-4°N). In the scalp the proportion of follicles in anagen reached a single peak of over 90% in March, and fell steadily to a trough in September. The number of shed hairs reached a peak around August/Septemher. when least follicles were in anagen. At this time the average loss of hairs was about 60 per day. more than double that during the preceding winter. The rate of growth of the beard was lowest in January and February and increased steadily from March to [uly to reach a peak about fiO% above the winter level. The rate of growth of thigh hair showed a similar pattern though with less pronounced differences. No seasonal fluctuations in finger- or toenail growth were detected.
The human species. Homo sapiens, has lost most of its body hair during the course of evolution. Conspicuous hair remains on the scalp, the axilla and pubic region and in the male beard. Elsewhere, except on the friction surfaces, hair follicles are present, but for the most part they produce only short and often barely visible hairs. Nevertheless, as in all other mammals the activity of all follicles is cyclifal. Phases of activity or atuujen are separated by phases of quiescence or tehcu'ii during which the fully grown hair remains anchored in the skin by an expanded base known as a club.'^ The club hair is shed after a new period of anagen has started. Scalp follicles may each remain active for several years, though more than 10% of them are resting at any particular time and 100 hairs or more may be shed each day from a single scalp.' In the thigh, in contrast, the mean duration of anagen has been estimated at only 54 days in males and less than half that In females.•* The cyclical activity of the follicle is the mechanism by which mammals change their pelage to meet the exigencies of growth, seasonal changes in the ambient environment and. perhaps, normal wear and tear."" Some species (for example, the Norway rat)'' grow new hair and moult in a patterned sequence. In others, such as the guinea-pig, the moults are more diffuse. Moulting appears to be controlled by an interaction of endogenous and exogenous factors. On the one hand, follicles have an endogenous rhythm which continues even when they are transplanted to other sites or other individuals. On the other hand, their cycles can be markedly accelerated or retarded by systemic factors."" It 146
seems likely that environmental cues, mainly the photoperiod but with some influence of temperature, mediate through the hypothalamo-hypophysial pathway the production of steroid and thyroid hormones and that these in turn influence the hair growth cycles. Human postpartum alopecia, which appears to be an interference with the hair cycle by hormones circulating in late pregnancy,** may be a legacy of this mechanism, as may diffuse alopecia linked with malfunction of the thyroid.'* Hair which is used for social or sexual communication, either as a visual signal or as a component of scent glands, is under separate control, for the adaptive reason that socio-sexual signals are a matter only for adults, not for immature animals still under parental care. Such structures are almost invariably under the influence of androgens:'" the growth of human axillary and pubic hair in both sexes, as well as the male beard and any other conspicuous hair on the male trunk and limbs, and sometimes also on the female, is similarly androgendependent. The fact that human hair lollicles. just as those of other mammals, undergo cyclical activity and are influenced by hormones raises a question of considerable evolutionary interest as well as of practical importance in the assessment of pharmacohjgical effects; does human hair growth fluctuate seasonally? A single report that three women in New York experienced maximum hair fall in November" .suggests this possibility. To provide a detinite answer, various parameters were determined over 18 months in 14 males.
SEASONAL CHANGES IN HAIR GROWTH
Methods Subjects
The subjects consisted of 14 healthy Caucasian men iiged 18-59 years with indoor occupations in Sheffield, U.K. (latitude 5 3 4°N). With the exception of one highlymotivated engineer, they ,were scientists or technicians (i.e. a particularly reliable, highly trained study group). They each collected beard shavings, shed head hair, finger- and toenail clippings and recorded the number of hours spent outdoors. They were visited every 2S days for 18 months: at each visit hair samples were obtained, by the same scientist, from five areas of the scalp and the thigh, an androgen-sensitive area. Methods h;id to be used which were both logistically feasible and likely lo produce data which could be rigorously analysed. Moreover, it was essential that ail procedures should be sufficiently non-invasive as to be repeatable and acceptable to the volunteer participants over a long period.
147
weighed. Wet shaved hair was first washed, filtered and dried. Ttmih Percentaife of follicles in athujen. At the first visit, a
template was used to delimit an area of 40 x 40 mm on the ventral side of the left thigh, and the corners were marked using a ball-point pen. The total number of hairbearing Ibllicles was then counted, before the whole area was shaved with a safety razor. To enable repetitive examination of the same area, a helghtmeasuring device was used to obtain vertical and horizontal coordinates for each subject. At the next examination the number of reappearing hairs was counted. Although this docs not directly refiect the number ttf follicles active at the moment of examination, it is a reasonable measureof any changes over the period, as the inclusit)n of club hairs in follicles which have ceased activity will be at least partly offset by the omission of follicles which have started activity but not yet grown identifiable emergent hair.
Scalp Percentage of follicles in anagen. In the scalp, the propor-
tion of follicles in anagen was assessed from the roots of plucked hairs. However, rather than sample hairs being pulled out as a mass from a single region, the so-called trichogram.'-' we plncked single, long hairs at random in each specific scalp area. Kach hair was gripped with spade-ended forceps placed as close as possibie to the skin and extracted by a sharp puli. Ten hairs were removed from the vertex, the left and right parietai regions, and the left and right occipital regions, making a total of SO hairs for each subject at each examination. Hair-root status was assessed under the dissecting microscope.'' Shed hairs. Subjects coilected ail hairs shed during combing or brushing over the 4 days prior to each visit. In addition, they were instructed to shampoo the scalp on the penultimate day. retaining all detached hairs using a piughoie sieve. Hair diameter. The hair diameter for each hair was measured at 7 and 14 mm from the base of the plucked root after careful placing on its widest axis.
Rate of ({rowth. At the second and subsequent visits 40 hairs judged to be among the longest, and thus the most likely to have been growing lor all or most oi the period, were removed at skin level using curved scissors, and measured.'' The area was then again shaved. lUiir diameter. The removed hairs were mounted in Canada balsam and their mean diameter measured.'"* Nails
Subjects were asked to save all nail parings from fingers and toes. Statistical procedures
1 he means and standard errors for each parameter for each month were calculated. In addition, the results were statistically assessed by non-parametric runs, turning points and /--tests.'"'
Results Heard
Outdoor artivitu
Subjects were asked to save all hair shaved from the face. by whatever method, during the 7 consecutive days immediately before each visit, and the samples were
Despite their indoor occupations the subjects sh(5wed pronounced seasonal differences in their behaviour (runs test < 0 ( ) l : turning points
ADONIS OOO7O96391OOO33Q
Seasonal changes in human hair growth VALER[K A.RANDALL AND F.).G.EBLING* Department of Biomedical Sciences, University of Bradford *Sub-ikpartmenl of Dermatology, Academic Division of Medicine. University of Sheffield. Royal Haiiamshire Hospital U.K.
Sbeffieid.
Accepted for publication 24 August 1990
Summary
Various parameters of hair growth were determined every 28 days for 18 months in 14 healthy Caucasian men aged 18-39 with indoor occupations in Sheffield, U.K. (latitude 5 1-4°N). In the scalp the proportion of follicles in anagen reached a single peak of over 90% in March, and fell steadily to a trough in September. The number of shed hairs reached a peak around August/Septemher. when least follicles were in anagen. At this time the average loss of hairs was about 60 per day. more than double that during the preceding winter. The rate of growth of the beard was lowest in January and February and increased steadily from March to [uly to reach a peak about fiO% above the winter level. The rate of growth of thigh hair showed a similar pattern though with less pronounced differences. No seasonal fluctuations in finger- or toenail growth were detected.
The human species. Homo sapiens, has lost most of its body hair during the course of evolution. Conspicuous hair remains on the scalp, the axilla and pubic region and in the male beard. Elsewhere, except on the friction surfaces, hair follicles are present, but for the most part they produce only short and often barely visible hairs. Nevertheless, as in all other mammals the activity of all follicles is cyclifal. Phases of activity or atuujen are separated by phases of quiescence or tehcu'ii during which the fully grown hair remains anchored in the skin by an expanded base known as a club.'^ The club hair is shed after a new period of anagen has started. Scalp follicles may each remain active for several years, though more than 10% of them are resting at any particular time and 100 hairs or more may be shed each day from a single scalp.' In the thigh, in contrast, the mean duration of anagen has been estimated at only 54 days in males and less than half that In females.•* The cyclical activity of the follicle is the mechanism by which mammals change their pelage to meet the exigencies of growth, seasonal changes in the ambient environment and. perhaps, normal wear and tear."" Some species (for example, the Norway rat)'' grow new hair and moult in a patterned sequence. In others, such as the guinea-pig, the moults are more diffuse. Moulting appears to be controlled by an interaction of endogenous and exogenous factors. On the one hand, follicles have an endogenous rhythm which continues even when they are transplanted to other sites or other individuals. On the other hand, their cycles can be markedly accelerated or retarded by systemic factors."" It 146
seems likely that environmental cues, mainly the photoperiod but with some influence of temperature, mediate through the hypothalamo-hypophysial pathway the production of steroid and thyroid hormones and that these in turn influence the hair growth cycles. Human postpartum alopecia, which appears to be an interference with the hair cycle by hormones circulating in late pregnancy,** may be a legacy of this mechanism, as may diffuse alopecia linked with malfunction of the thyroid.'* Hair which is used for social or sexual communication, either as a visual signal or as a component of scent glands, is under separate control, for the adaptive reason that socio-sexual signals are a matter only for adults, not for immature animals still under parental care. Such structures are almost invariably under the influence of androgens:'" the growth of human axillary and pubic hair in both sexes, as well as the male beard and any other conspicuous hair on the male trunk and limbs, and sometimes also on the female, is similarly androgendependent. The fact that human hair lollicles. just as those of other mammals, undergo cyclical activity and are influenced by hormones raises a question of considerable evolutionary interest as well as of practical importance in the assessment of pharmacohjgical effects; does human hair growth fluctuate seasonally? A single report that three women in New York experienced maximum hair fall in November" .suggests this possibility. To provide a detinite answer, various parameters were determined over 18 months in 14 males.
SEASONAL CHANGES IN HAIR GROWTH
Methods Subjects
The subjects consisted of 14 healthy Caucasian men iiged 18-59 years with indoor occupations in Sheffield, U.K. (latitude 5 3 4°N). With the exception of one highlymotivated engineer, they ,were scientists or technicians (i.e. a particularly reliable, highly trained study group). They each collected beard shavings, shed head hair, finger- and toenail clippings and recorded the number of hours spent outdoors. They were visited every 2S days for 18 months: at each visit hair samples were obtained, by the same scientist, from five areas of the scalp and the thigh, an androgen-sensitive area. Methods h;id to be used which were both logistically feasible and likely lo produce data which could be rigorously analysed. Moreover, it was essential that ail procedures should be sufficiently non-invasive as to be repeatable and acceptable to the volunteer participants over a long period.
147
weighed. Wet shaved hair was first washed, filtered and dried. Ttmih Percentaife of follicles in athujen. At the first visit, a
template was used to delimit an area of 40 x 40 mm on the ventral side of the left thigh, and the corners were marked using a ball-point pen. The total number of hairbearing Ibllicles was then counted, before the whole area was shaved with a safety razor. To enable repetitive examination of the same area, a helghtmeasuring device was used to obtain vertical and horizontal coordinates for each subject. At the next examination the number of reappearing hairs was counted. Although this docs not directly refiect the number ttf follicles active at the moment of examination, it is a reasonable measureof any changes over the period, as the inclusit)n of club hairs in follicles which have ceased activity will be at least partly offset by the omission of follicles which have started activity but not yet grown identifiable emergent hair.
Scalp Percentage of follicles in anagen. In the scalp, the propor-
tion of follicles in anagen was assessed from the roots of plucked hairs. However, rather than sample hairs being pulled out as a mass from a single region, the so-called trichogram.'-' we plncked single, long hairs at random in each specific scalp area. Kach hair was gripped with spade-ended forceps placed as close as possibie to the skin and extracted by a sharp puli. Ten hairs were removed from the vertex, the left and right parietai regions, and the left and right occipital regions, making a total of SO hairs for each subject at each examination. Hair-root status was assessed under the dissecting microscope.'' Shed hairs. Subjects coilected ail hairs shed during combing or brushing over the 4 days prior to each visit. In addition, they were instructed to shampoo the scalp on the penultimate day. retaining all detached hairs using a piughoie sieve. Hair diameter. The hair diameter for each hair was measured at 7 and 14 mm from the base of the plucked root after careful placing on its widest axis.
Rate of ({rowth. At the second and subsequent visits 40 hairs judged to be among the longest, and thus the most likely to have been growing lor all or most oi the period, were removed at skin level using curved scissors, and measured.'' The area was then again shaved. lUiir diameter. The removed hairs were mounted in Canada balsam and their mean diameter measured.'"* Nails
Subjects were asked to save all nail parings from fingers and toes. Statistical procedures
1 he means and standard errors for each parameter for each month were calculated. In addition, the results were statistically assessed by non-parametric runs, turning points and /--tests.'"'
Results Heard
Outdoor artivitu
Subjects were asked to save all hair shaved from the face. by whatever method, during the 7 consecutive days immediately before each visit, and the samples were
Despite their indoor occupations the subjects sh(5wed pronounced seasonal differences in their behaviour (runs test < 0 ( ) l : turning points
