Methods of evaluating hair growth

Australasian Journal of Dermatology

(2003)

44

,

10–18

Correspondence: Dr Alexander J Chamberlain, Clinical ResearchFellow, Department of Dermatology, Churchill Hospital, Old Road,Headington, Oxford OX3 7LJ, UK. Email: [email protected]

Alexander J Chamberlain, MB BS. Rodney PR Dawber, FRCP.Submitted 27 March 2002; accepted 15 August 2002.

PERSONAL REVIEW

Methods of evaluating hair growth

Alexander J Chamberlain and Rodney PR Dawber

Department of Dermatology, Churchill Hospital, Headington, Oxford, United Kingdom

INTRODUCTION

Many methods of evaluating the growth of hair have beendescribed by both scientists and clinicians alike. While themajority of these methods have been developed to progressthe understanding of normal hair physiology, the discoveryof agents that promote hair growth has necessitated thedevelopment of reliable and reproducible means of assessinghair loss and monitoring response to therapy. Indeed, themajority of techniques developed to date have followed thediscovery of minoxidil and its application in androgeneticalopecia (AGA). Without this revolution, it is unlikely thatmuch of the intense hair growth research and innovation

over the last two decades would have occurred at all.Although sophisticated computerized techniques pioneeredby the wool industry are now available to quantify thevarious parameters of hair growth, the intrinsic complexityof hair growth itself has been the main obstacle to thedevelopment of simple office-based assessment tools. Theaim of this review is to outline the various methods availablefor the measurement of hair growth, their limitations andtheir value to the clinician.

THE HAIR CYCLE

Much of our understanding of hair growth is based onanimal studies.

1

The dynamics of the hair cycle not onlydiffer between species and race but also with age, gender,body site and across seasons. Physiological or pathologicalstates such as pregnancy, malnutrition or malignancy canalso modulate hair growth. The total number of follicularunits on the human scalp is approximately 100 000–150 000and it is generally accepted that this number is fixed anddoes not increase. These follicles are quite evenly distributedover the scalp and most typically in groups of three. Althoughthese small groups grow in phase, the growth of scalp hairon the whole is asynchronous or mosaic. Hair folliclesundergo active growth (anagen) in a cyclic fashion, with thisphase typically lasting 3–7 years on the scalp vertex. Anagenhas traditionally been divided into six stages of development,from the onset of mitotic activity within the secondary hairgerm (anagen-I) to the fully developed follicle that firstemerges beyond the skin surface (anagen-VI). After a brieftransitional phase (catagen) hairs enter a resting phase(telogen) lasting weeks before being shed. This process thenrepeats itself over and over. At any one time, 85–90% of scalphairs are in anagen, with the majority of the remainder intelogen. The growth rate of the hair follicle is fairly constant,averaging about 0.5 mm per day on the scalp vertex and alittle less at the margins.

DIMENSIONS OF HAIR GROWTH

It is important to define the dimensions that constitute hairgrowth before one can begin to interpret the various methodsof its measurement. The basic biological parameters of hairgrowth include rate of linear growth, hair shaft diameter,hair density and hair-cycle status. The anagen : telogen ratio(or percentage of anagen-VI hairs) is the best measure ofcycle status given the impracticalities of measuring theduration of anagen, which relies on long periods of

SUMMARY

For decades, scientists and clinicians have examinedmethods of measuring scalp hair growth. With thedevelopment of drugs that stem or even reverse theminiaturization of androgenetic alopecia, there hasbeen a greater need for reliable, economical andminimally invasive means of measuring hair growthand, specifically, response to therapy. We review thevarious methods of measurement described to date,their limitations and value to the clinician. In ouropinion, the potential of computer-assisted technologyin this field is yet to be maximized and the currentlyavailable tools are less than ideal. The most valuablemeans of measurement at the present time are globalphotography and phototrichogram-based techniques(with digital image analysis) such as the ‘TrichoScan’.Subjective scoring systems are also of value in theoverall assessment of response to therapy and theseare under-utilized and merit further refinement.

Key words: androgenetic alopecia, epiluminescencemicroscopy, phototrichogram, scalp photography,trichogram.

Methods of evaluating hair growth 11

observation and identification of individual hairs. Pigmentcontent must also be considered a variable of some biologicalsignificance in the study of alopecia.

It is crucial to recognize that the parameters of hair growththat are of major interest to the scientist are different to thoseof the clinician or the patient. While much of the recentscientific research in this area has concentrated on themechanisms that control the activity of putative follicularstem cells, the parameters more likely to be of value toclinicians are diameter, density and cycle status. In contrast,the primary concern of the patient is reduction in density,which is frequently expressed as ‘loss of body’. The patientwho is losing hair may also report an increase in shedding(‘hair is falling out’), a reduction in growth rate (‘hair doesn’tneed cutting as often’), or a reduction in diameter (‘hair isfiner’). Of course, the presentation of AGA is quite differentin women and men. Unlike a woman with diffuse hair lossin whom the diagnosis may not be immediately clear, a manpresenting with a patterned alopecia is usually all toofamiliar with ‘common baldness’ and presents requestingtreatment rather than a diagnosis.

At present, there is no ‘gold standard’ in the area ofmeasurement of hair growth. The ideal measure would benon-invasive, simple to perform, reproducible, economical,and able to integrate all of the basic biological parameters ofhair growth as well as to provide information that ismeaningful to the patient. The vast majority of the methodsof measurement that will be discussed here have beenstudied in AGA.

These methods of measurement are equally applicable toother non-scarring alopecias, with the exception of patchyalopecia areata, in which areas of hair loss are well-definedand regrowth is easily identified; therefore accuratemeasurement is not as crucial to the clinician. In scarringalopecias, histology is vital for diagnosis but quantitativeestimates of hair loss are again not as important. In bothalopecia areata and scarring alopecia, clinical examinationand global photography provide the information of greatestrelevance, although a more detailed discussion of this issuecan be found elsewhere and is beyond the scope of thisreview. The ideal frequency of measurement in the monitor-ing of hair growth in response to therapy is another factor inthe equation that is yet to be established. On the basis oftelogen duration, many clinical trials have chosen tomeasure

response

to

therapy

at

4-monthly

intervals,

andthis

is

probably

the

minimum

interval

at

which

changesmay

be

detected.

As

any

expected

changes

in

hairparameters will be small if at all, it is probably prudent toassess at 6–12-monthly intervals until a trend has beenestablished. The various methods of measurement of hairgrowth have been previously divided into non-invasive,semi-invasive and invasive categories and this is a logicaldivision that we have also chosen to structure our discussionaround.

NON-INVASIVE METHODS

Only recently has a validated and standardized question-naire aimed at the detection of subjective changes in hair

growth in men with AGA been developed.

2

This question-naire was developed in the Merck Research Laboratories andhas already been used to supplement clinical measures intrials examining the effects of finasteride.

3

The questionnaireconsists of five questions that address patient perceptions ofhair growth and satisfaction with hair appearance. Thescores have been shown to correlate with hair counts, albeitmodestly. This subjective measure is probably a usefuladjunct in the assessment of response to therapy. While thisquestionnaire was designed to assess the perceptions of men,it would not be difficult to adapt this so that it were applicableto women also.

The manual collection, counting and weighing of hairsshed on a daily basis is a simple technique that can be usedto follow the progression of hair loss.

4

Studies of this typehave estimated that the average number of hairs shed dailyis about 100, a figure that appears widely in hair literatureand is probably an overestimate for healthy young adults.This crude method is not only extremely tedious but requirescareful standardization of collection technique and hair care.Hydration and seborrhoea are factors that affect hair weight,so these must also be standardized if this technique is to beapplied in clinical trials.

There are a number of grading scales that exist for theclassification of AGA. It was Hamilton in 1951 who firstsubdivided the patterning seen in men with AGA into typesI–VIII on a visual scale.

5

This system of classification waslater modified by Norwood in 1975 who added four furtherintermediate grades: IIIa, III vertex, IVa and Va.

6

Althoughthe Norwood–Hamilton scale has been used extensively inclinical trial evaluation, this system is too crude to reliablydocument response to therapy. In reality, the progression ofAGA or indeed regrowth with therapy occurs over a con-tinuum rather than in a stepwise manner. This scale prob-ably still has a role in the broader classification of severity ofAGA in men, particularly in the identification of those whoare more likely to respond to therapy.

In 1977, Ludwig proposed a grading scale for women withAGA.

7

He suggested types I–III that represented a range ofdiffuse alopecia over the crown in the so-called ‘femalepattern’ that spares the frontal hair line. It is well recognizedthat a certain percentage of women develop AGA in the ‘malepattern’, particularly postmenopausal women.

8

Savin has expanded on the original grading systems withhis pattern and density scales for classifying AGA in bothmen and women (Fig. 1).

9

In men, the appearance of thethinning

frontal

scalp,

mid

area

and

vertex

are

rankedF1–F6, M1–M5 and V1–V7, respectively. Female-pattern AGAis divided according to the width of the central part. Densityin both sexes is also graded on the basis of the part (D1–D8).Unfortunately, these qualitative systems are too crude to usein clinical trials for anything more than defining the base-line. They probably have a role in office-based measurementuntil more sensitive tools are widely available and they arecertainly significant from a historical perspective.

Photography is a very important tool in the clinicalmeasurement of hair and its dynamics. Global photographsof the scalp are clinically relevant in the assessment of AGAbecause they represent an accurate record of the cosmetic

12 AJ Chamberlain and RPR Dawber

state of the patient. High-quality photographic systems withstereotactic positioning devices for accurate patient position-ing are now available commercially.

10

Factors such asprecise combing and lighting are critical to the repro-ducibility of scalp photographs. The Canfield technique hasrecently been validated in a randomized controlled trialexamining men with Hamilton III or IV AGA receiving oralfinasteride or placebo.

11

Four standard global views (vertex,midline, frontal and temporal) are advocated, although only

vertex views were used in the early finasteride trials (Fig. 2).

3

It is our opinion that global photographs are a valid methodof hair measurement in the clinical setting, although the bestresults are probably obtained in the hands of trained medicalphotographers.

In recent years, epiluminescence microscopy (ELM) hasbeen cleverly adapted to quantify hair growth. Epilumines-cence microscopy is a clinical tool that was originally refinedfor the diagnosis of pigmented skin lesions, in particular theearly detection of melanoma. An European group hascreated a scoring system based on hair density and diameteras assessed by macrophotographs of a midline part at thevertex

of

men

with

AGA

(Fig. 3).

12

These

are

performedusing a commercially available camera (Dermaphot; HeineOptotechnik, Herrsching, Germany) designed for ELM with

�

4 magnification. An area measuring 14

�

13 mm is used tobase the measurements. Density is graded on a scale of 1(fewer than four hairs) to 6 (more than 40 hairs) and dia-meter is graded as 1 (thin), 2 (medium) or 3 (thick). Aseparate hair diameter diversity scale is graded 0 (<20% inhair diameter diversity) or 1 (>20% hair diameter diversity).They propose that a score of >20% diversity in hair diameteris a useful clinical sign reflecting miniaturization. This istherefore of potential use in the diagnosis of AGA, but prob-ably too crude for monitoring change. Their density anddiameter scale has been shown to correlate well with theHamilton classification and histological hair density. In someways it can be considered an adaptation of the Savin scale,but localized to a precise point on the vertex and undermagnification. A key issue to consider is whether a postagestamp-sized area of scalp vertex is representative of theglobal effects of AGA on the scalp. Even if it is, this techniqueis still qualitative and quite subjective, but easy to performand non-invasive.

Figure 1

Pattern and density Savin scales.

Figure 2

Schematic representation of stereotactic scalp photo-graphy demonstrating fixation of the subject on a brow and chin restwith the camera mounted on a rotating arm.


The information technology revolution of the late twen-tieth century has resulted in advances throughout medicine,including hair measurement. The major capability of thecomputer that has been exploited in the field of hairmeasurement is image analysis and its quantification ofclinical or even histological appearances. High-speedprocessing, high-resolution imaging and vast storage cap-acity for archiving are further advantages of computersystems in clinical measurement. A computer-assistedmethod

of

estimating

hair

density

has

been

used

toestimate the response to therapy in a patient with diffusealopecia areata.

13

This involves the digitization of a photo-graphic image of the scalp and could equally be applied toAGA. Darker shades of grey are interpreted as hair andlighter shades as scalp and then the proportion of eachvariable is calculated. Factors that influenced the results

were hair styling and combing, camera-to-scalp distance andcontrast between scalp and hair colour. These are short-comings

that

the

Canfield

system

has

partly

addressedwith standardized combing and patient positioning, althoughthe contrast between scalp and hair colour, particularlywhen it is minimal, is a limitation common to all imagingsystems.

SEMI-INVASIVE METHODS

The linear growth rate of scalp hair can be measured in anumber of ways. Perhaps the simplest technique is to stain agroup of hairs using a dye that contrasts with the colour ofthe native hairs.

14

A growth rate can then be calculated bydividing the length of unstained hair in millimetres by thenumber of days between measurements (mm/day). If

Figure 3 Scoring system of hairdensity and shaft diameter using amagnified photographic scale (repro-duced with permission from Arch.Dermatol. 2001; 137: 641–6, copy-righted 2001, American MedicalAssociation).


estimated over a standardized area, the proportion of anagenand telogen hairs can also be assessed, because only theanagen hairs will show growth. Close-up photographyimproves the accuracy of such observations, although con-ditions must be strictly standardized and there must besufficient contrast between background scalp and haircolour.

15

Alternatively, hairs may be clipped close to the surface,shaved or even plucked prior to measuring growth rate,although plucking results in a variable delay before theappearance of a new anagen hair. A designated area can bere-clipped, shaved or plucked and then weighed after a setperiod to measure regrowth. This technique has been usedto document normal growth,

16

and response to minoxidil inwomen

17

and finasteride in men

18

with AGA. It has beenproposed that the total weight of hair from a defined area ofscalp is an ideal quantitative measure of hair growth for usein clinical trials.

17

However, this technique seems just ascomplex and tedious as other manual methods, requiringcareful degreasing, drying and control of humidity to avoiderror.

Calibrated capillary tubes have also been used to measurelinear growth rate.

19

These are fitted around individual hairsthat have been clipped and then pressed gently against theskin surface (Fig. 4). Measurements are daily or even morefrequently. Reference points would clearly be necessary ifparticular hairs were to be followed individually over time.Intradermal injection of radiolabelled substances followedby autoradiography is a research technique that provides anaccurate method of marking hairs along their length.

20

Linear growth rate can then be measured by dividing thedistance between marks on autoradiographs by the timebetween injections. The small doses of radiation required inthese assays have been deemed safe in non-pregnanthumans. These techniques are not only labour-intensive butserial scalp injections are unlikely to be well tolerated bypatients outside a research setting. Linear growth rate is nota vital parameter in the measurement of miniaturization.

In the 1960s a compound measurement of the majorbiological parameters of hair growth was first described.

21

This technique was to be known later as the trichogram.

22

This involved shaving a small area of scalp hair as well asplucking 50–100 hairs using forceps without sharp edges andideally covered with adhesive tape (Fig. 5). The shaved areawas examined after 10 days to assess the linear growth rate.The roots of the plucked hairs were examined to provide anestimate of cycle status (anagen : telogen differential) anddensity was assessed using a microscope with

�

60 magnifi-cation and an eyepiece marked with a 1-cm

2

area. Hairdiameter was also assessed microscopically. Originally, hairswere graded into thick (0.1 mm), medium (0.05 mm) andthin (0.025 mm) categories using calibrated wires forcomparison. A final parameter was the ‘regeneration period90’, which was the time lag in the appearance of 90% ofplucked hairs from a given area. This technique was appliedextensively thereafter to study the hair of subjects of varyingages and under various physiological influences.

23

While thiswork helped to define the normal variations according toage, gender and body site, it is now largely outdated as aresearch technique in AGA. A further refinement on thetrichogram was the unit area trichogram (UAT), whichintroduced a standardized pre-sampling, washing andcombing protocol, and plucking of hairs from within adefined area (usually 35–44 mm

2

) to determine density in amore accurate fashion.

24

The major shortfall of thetrichogram is that hairs in early anagen and vellus hairs areeasily missed in a standard pluck because of their small size.Furthermore, plucking 50–100 hairs is quite a painful pro-cedure for patients to undergo. Plucking is also known tochange the natural course of the hair cycle.

25

The first efforts to document the growth of hair usingphotography at close range were performed on threeJapanese men who were studied at regular intervals over aperiod of 2 years.

15

Hairs at different body sites including thescalp were studied, enabling estimates of the lengths ofanagen and telogen for each individual. Hairs were keptclipped at lengths of 1 mm (whenever photographed) withina small area to allow accurate identification of individualhairs.

Figure 4

Hair-growth measurement using graduated scalecapillary tubes.

Figure 5

Technique for plucking hairs with surgical forceps.


This work formed the scientific basis of the photo-trichogram (PTG) that has been refined by several investi-gators.

26,27

This technique involves serial magnifiedphotographs (Fig. 6) of a target area of scalp hair (1 cm

2

) overa period of days to define which hairs are in anagen (thosegrowing) and which are in telogen (those not growing). Inaddition to cycle status, the PTG enables measurement oflinear growth rate as well as diameter and density. Althoughpainless, it is time-consuming and is limited by the sameinherent problem as other scalp imaging systems ofdifferentiating hairs from scalp background. The applicationof an immersion oil

28

and temporary hair dye

29

haveimproved this contrast to some degree. Several investigatorshave attempted to automate the process of monitoring hairgrowth via PTG, although overlapping hairs and variousartefacts have proven to be major obstacles to computerizedmeasurement.

27,30,31

Assessment of hair diameter is alsounreliable unless magnifications in the order of

�

20 areused.

32

A computer-assisted image analyzer has been shownto enhance the quality and reliability of this facet ofmeasurement.

33

Despite these shortfalls, the PTG has been applied success-fully in finasteride trials to show statistically significantincreases in density and the anagen : telogen ratio over12 months in men with AGA.3,34 Increases in hair diameterhave not been able to be adequately documented in thesetrials using the PTG.

The latest modification of the PTG combines ELM withdigital image analysis; the ‘TrichoScan’.35 In comparison tothe traditional PTG, this method uses ELM instead ofmacrophotography, as well as contrast enhancement of hairswith temporary dye. Specially developed computer softwareenables assessment of the four basic biological parameters ofhair growth (Fig. 7), with the whole procedure taking20 min. This technique has already been used to documentincreases in hair count and cumulative diameter in men withAGA treated with finasteride, over a period of 6 months. Thedevelopers of the TrichoScan appear to have drawn on yearsof PTG research to create a validated clinic-based measure-ment tool that is both elegant and patient friendly. The costof equipping the clinic with this system is an importantoutstanding question.

INVASIVE METHODS

The ultimate measure of hair growth is the examination ofmatrix cell kinetics. The two main proliferative indices arethe mitotic index and the labelling index.36 The former is acount of the number of actively dividing cells at a given pointin time, as measured on wax-embedded sections. The latteris a count of the number of cells entering mitosis during agiven period, and requires intradermal injections of tritiatedthymidine prior to biopsies subjected to light microscopyautoradiography at set intervals.37 These indices are not onlyinvasive, but they only provide a static appreciation of adynamic process. In addition, thymidine is potentially toxicand therefore unsuitable for repeated administration. Themetaphase index, in contrast, provides a measure of the rateof cell production and is more suitable for studying hair

growth. Cells arrested in metaphase are expressed as aproportion of the entire matrix. This technique has mainlybeen applied to the study of the seasonal variations in woolgrowth.38 However, this also involves administration ofcytotoxics.

Histopathological analysis of scalp biopsies has been usedfor some time now to investigate hair growth. Biopsies areperformed under local anaesthesia. A 4- or 6-mm punchdeep into the subcutaneous tissue directed along the slant ofexiting hairs is typically taken from a transitional areabetween balding and normal scalp and a suture is normallyrequired for haemostasis. Although the original descriptionsof this technique involved a single punch, it is more usual totake two punches at the present time. Early studies ofmacaque monkeys (who develop AGA) using scalp biopsiessectioned vertically along the length of the hair follicleenabled the construction of histograms known as ‘folliculo-grams’.39 These depicted follicle lengths and proportions ofanagen, catagen and telogen hairs, and over time were ableto demonstrate the improvement in the terminal : vellus ratiowith minoxidil therapy.40

It has since been shown that horizontal sectioningprovides greater diagnostic information than vertical sec-tions alone.41 This mode of processing provides measures ofdensity, hair-shaft diameter, and the anagen : telogen andterminal : vellus ratios. Automated computer analysis hasbeen successfully adapted to assist with the quantification ofthese parameters.42 Both density and terminal : vellus ratioshave been shown to be useful indices in monitoring theresponse of AGA to minoxidil43 and finasteride,44 respec-tively. A terminal : vellus ratio of less than 4:1 has beensuggested as diagnostic of AGA, whereas the ratio in the

Figure 6 Macrophotograph of scalp hair.


normal scalp is between 6:1 and 8:1.43 One of the criticismsof this method is that horizontal sections taken at the level ofthe entry of the sebaceous duct will underestimate the totalnumber of vellus hairs because many of these are foundmore superficially in the midreticular dermis.38

Another histological parameter that has been studied is thevolume of the hair bulb. In animal studies the volume of thedermal papilla has been shown to bear a direct relationshipto the volume of terminal hair.45 Although this work has beenreplicated in humans with AGA,46 this parameter has notbeen used to follow the response of AGA to therapy.

The major limitation of these invasive methods in thestudy of hair growth is that they do not permit the repeatedsampling of a consistent area of scalp over time. Further-more, a single 4-mm sample of scalp is not alwaysadequately representative of the global process because ofregional variation.47

CONCLUSION

An adequate means of measuring hair growth in the clinicover time in a reproducible, economical and non-invasive

manner does not appear to be available at the presenttime, despite the intense research into this subject. Thelarge variety of techniques employed over the decades istestament to this conclusion. The ideal tool(s) would beable to measure hair-shaft diameter and hair density at amicroscopic level, in addition to providing a subjectiveassessment of both the appearance and satisfaction of thepatient.

While the UAT and the PTG may be the favoured toolswithin the setting of clinical trials, we feel that subjectivescores and global photography are of greater relevance in theclinic. A subjective scoring system for women with AGA oreven a unisex scoring system are yet to be developed andvalidated as far as we are aware. In our opinion, the methodthat shows the greatest promise in the monitoring of AGAand its response to therapy is the TrichoScan, although thisis only likely to be financially viable for academic depart-ments (as opposed to the average clinician) at the presenttime. While the impact of computer-assisted image analysishas been largely disappointing as far as progressingtechniques of hair measurement, we hope that the nextdecade will see the introduction of improved means of

Figure 7 Software interface depicting variables of hair growth as measured by the ‘Trichoscan’ technique (reproduced with permission fromEur. J. Dermatol. 2001; 11: 362–8).


assessing patients with AGA, ideally adaptable for otherdisorders of hair growth.

ACKNOWLEDGEMENTS

Dr Chamberlain was the recipient of the 2001 NovartisTravelling Fellowship and wishes to gratefully acknowledgeNovartis Australia and the Australasian College of Derma-tologists for their support. We also wish to thank OxfordMedical Illustration at the John Radcliffe Hospital for theirtechnical assistance with the artwork.

REFERENCES

1. Chase HB. Growth of the hair. Physiol. Rev. 1954; 34: 113–26.2. Barber BL, Kaufman KD, Kozloff RC, Girman CJ, Guess

HA. A hair growth questionnaire for use in the evaluationof therapeutic effects in men. J. Dermatol. Treat. 1998; 9:181–6.

3. Kaufman K, Olsen EA, Whiting D, Savin R, DeVillez R,Bergfeld W et al. Finasteride in the treatment of men withandrogenetic alopecia. J. Am. Acad. Dermatol. 1998; 39: 578–89.

4. Orentreich N. Scalp hair replacement in man. In: Montagna W,Dobson RL (eds). Advances in Biology of the Skin, Vol. IX.Oxford: Pergamon Press, 1969; 99–108.

5. Hamilton JB. Patterned loss of hair in man: Types and incidence.Ann. NY. Acad. Sci. 1951; 53: 708–28.

6. Norwood O’TT. Male pattern baldness. Classification and inci-dence. South. Med. J. 1975; 68: 1359–70.

7. Ludwig E. Classification of the types of androgenetic alopecia(common baldness) occurring in female sex. Br. J. Dermatol.1997; 97: 247–54.

8. Venning VA, Dawber RPR. Patterned androgenic alopecia inwomen. J. Am. Acad. Dermatol. 1988; 18: 1073–7.

9. Savin RC. A method for visually describing and quantitating hairloss in male pattern baldness (Abstract). J. Invest. Dermatol.1992; 98: 604.

10. Canfield D. Photographic documentation of hair growth inandrogenetic alopecia. Dermatol. Clin. 1996; 14: 713–21.

11. Kaufman K, Binkowitz B, Savin R, Canfield D. Reproducibility ofglobal photographic assessments of patients with male patternbaldness in a clinical trial with finasteride (Abstract). J. Invest.Dermatol. 1995; 104: 659.

12. De Lacharriere O, Deloche C, Misciali C, Piraccini BM, VincenziC, Bastien P, Tardy I, Bernard BA, Tosti A. Hair diameterdiversity. A clinical sign reflecting the follicle miniaturisation.Arch. Dermatol. 2001; 137: 641–6.

13. Gibbons RD, Fiedler-Weiss VC, West DP, Lapin G. Quantificationof scalp hair – a computer-aided methodology. J. Invest.Dermatol. 1986; 86: 78–82.

14. Bartosova L, Jorda V, Stava Z. Laboratory and experimentaltrichology. In: Mali JWH (ed.). Diseases of the Hair and Scalp.New York: Karger, 1984; 224–38.

15. Saitoh M, Uzuka M, Sakamoto M. Human hair cycle. J. Invest.Dermatol. 1970; 54: 65–81.

16. Hamilton JB. Age, sex and genetic factors in the regulation ofhair growth in man. A comparison of Caucasian and Japanesepopulations. In: Montagna W, Ellis RA (eds). The Biology ofHair Growth. New York: Academic Press, 1958; 399–433.

17. Price V, Menefee E. Quantitative estimation of hair growth I.Androgenetic alopecia in women: Effect of minoxidil. J. Invest.Dermatol. 1990; 95: 683–7.

18. Price VH, Menefee E, Sanchez M, Ruane P, Kaufman KD.Changes in weight and hair count in men with androgeneticalopecia after treatment with finasteride, 1 mg, daily. J. Am.Acad. Dermatol. 2002; 46: 517–23.

19. Saitoh M, Uzuka M, Sakamoto M. Rate of hair growth. In:Montagna W, Dobson RL (eds). Advances in Biology of the Skin,Vol. IX. Oxford: Pergamon Press, 1969; 183–201.

20. Munro DD. Hair growth measurement using intradermalsulphur-35 cystine. Arch. Dermatol. 1966; 93: 119–22.

21. Barman JM, Pecoraro V, Astore I. Method, technique andcomputations in the study of the trophic state of human scalphair. J. Invest. Dermatol. 1964; 42: 421–6.

22. Barman JM, Astore I, Pecoraro V. The normal trichogram of theadult. J. Invest. Dermatol. 1965; 44: 233–6.

23. Pecoraro V, Astore IPL. Measurement of hair growth underphysiological conditions. In: Orfanos CE, Happle R (eds). HairDiseases. Berlin: Springer-Verlag, 1989; 237–54.

24. Rushton H, James KC, Mortimer CH. The unit area trichogramin the assessment of androgen-dependent alopecia. Br. J.Dermatol. 1983; 109: 429–37.

25. Dolnich EH. Variability in hair growth in Macaca mulatta. In:Montagna W, Dobson RL (eds). Advances in Biology of the Skin,Vol. IX. Oxford: Pergamon Press, 1969; 121–8.

26. Bouhanna P. The phototrichogram: A technique for theobjective evaluation of the diagnosis and course of diffusealopecias. In: Montagna W, Serri F, Bartoletti L, Celleno L,Morganti F, Secchi GF (eds). Hair and Aesthetic Medicine.Rome: Salus Internazionale, 1984; 277–80.

27. Pelfini C, Calligaro A. Some notes on the evaluation of hairgrowth by means of morphometric computerized analysis.J. Appl. Cosmetol. 1986; 4: 67–76.

28. Van Neste DJJ, Dumortier M, De Brouwer B, De Coster W. Scalpimmersion proxigraphy (SIP): An improved imaging techniquefor phototrichogram analysis. J. Eur. Acad. Dermatol. Venereol.1992; 1: 187–91.

29. Van Neste DJJ. Contrast enhanced phototrichogram (CE-PTG):An improved non-invasive technique for measurement of scalphair dynamics in androgenetic alopecia – validation study withhistology after transverse sectioning of scalp biopsies. Eur. J.Dermatol. 2001; 11: 326–31.

30. Van Neste D, Dumortier M, De Coster W. Phototrichogramanalysis: Technical aspects. and problems in relation to auto-mated quantitative evaluation of hair growth by computer-assisted image analysis. In: Van Neste D, Lachapelle JM,Antoine JL (eds). Trends in Human Hair Growth and AlopeciaResearch. Dordrecht: Kluwer Academic Publishers, 1989;155–65.

31. Hayashi S, Mitamoto I, Takeda K. Measurement of human hairgrowth by optical microscopy and image analysis. Br. J.Dermatol. 1991; 125: 123–9.

32. Rushton DH, De Brouwer B, De Coster W, Van Neste DJJ.Comparative evaluation of scalp hair by phototrichogram andunit area trichogram analysis within the same subject. ActaDerm. Venereol. 1993; 73: 150–3.

33. Tsuji Y, Ishino A, Hanzawa N, Uzuka M, Okazaki K, Adachi K,Imamura S. Quantitative evaluations of male pattern baldness.J. Dermatol. Sci. 1994; 7 (Suppl.): S136–41.

34. Van Neste D, Fuh V, Sanchez-Pedreno P, Lopez-Bran E,Wolff H, Whiting D et al. Finasteride increases anagen hair inmen with androgenetic alopecia. Br. J. Dermatol. 2000; 143:804–10.

35. Hoffman R, TrichoScan. Combining epiluminescence micro-scopy with digital image analysis for the measurement of hairgrowth in vivo. Eur. J. Dermatol. 2001; 11: 362–8.

36. Barth JH. Measurement of hair growth. Clin. Exp. Dermatol.1986; 11: 127–38.

37. Epstein WL, Maibach HI. Cell proliferation and movement inhuman hair bulbs. In: Montagna W, Dobson RL (eds). Advancesin Biology of the Skin, Vol. IX. Oxford: Pergamon Press,1969; 83–97.

38. Fraser IEB. Cell proliferation in the wool follicle bulb. In: LyneAG, Short BF (eds). Biology of the Skin and Hair Growth.Sydney: Angus and Robertson, 1965; 427–45.


39. Uno H, Adachi K, Montagna W. Morphological and biochemicalstudies of hair follicle in common baldness of stumptailedmacaque (Macaca speciosa). In: Montagna W, Dobson RL (eds).Advances in Biology of the Skin, Vol. IX. Oxford: Pergamon Press,1969; 221–41.

40. Uno H, Cappas A, Schlagel C. Cyclic dynamics of hair folliclesand the effect of minoxidil on the bald scalps of stumptailedmacaques. Am. J. Dermatopathol. 1985; 7: 283–97.

41. Headington JT. Transverse microscopic anatomy of humanscalp. Arch. Dermatol. 1984; 120: 449–56.

42. Lee MS, Kossard S, Wilkinson B, Doyle JA. Quantification of hairfollicle parameters using computer image analysis: A com-parison of androgenetic alopecia with normal scalp biopsies.Australas. J. Dermatol. 1995; 36: 145–7.

43. Whiting DA. Diagnostic and predictive value of horizontalsections of scalp biopsy specimens in male pattern androgeneticalopecia. J. Am. Acad. Dermatol. 1993; 28: 755–63.

44. Whiting DA, Waldstreicher J, Sanchez M, Kaufman KD.Measuring reversal of hair miniaturisation in androgeneticalopecia by follicular counts in horizontal sections of serialscalp biopsies: Results of finasteride 1 mg treatment of men andpost-menopausal women. J. Invest. Dermatol. 1999; 4 (Suppl.):S282–4.

45. Ibrahim L, Wright EA. A quantitative study of hair growth usingmouse and rat vibrissial follicles I. Dermal papilla volumedetermines hair volume. J. Embryol. Exp. Morphol. 1982; 72:209–24.

46. Alcaraz MV, Villena A, Perezde Vargas I. Quantitative study ofthe human hair follicle in normal and androgenetic alopecia.J. Cutan. Pathol. 1993; 20: 344–9.

47. Van Neste D. Human scalp hair growth and loss evaluationmethods: Is there simple and reliable method? Exp. Dermatol.1999; 8: 299–310.

Methods of evaluating hair growth

Health & Medicine