Morphometric Analysis of the Human Scalp Hair Follicle:Practical Implications for the Hair Transplant Surgeon and HairRegeneration Studies

FRANCISCO JIMENEZ, MD,� ANDER IZETA, PHD,y AND ENRIQUE POBLET, MDz

BACKGROUND The bulge stem cell region is a structure important for the regeneration of the pilose-baceous unit. Measurements of the different compartments of a hair follicle may have implications inhair transplantation and hair regeneration studies.

OBJECTIVE To measure the length of the different portions of the occipital scalp hair and to estimate atwhat depth they are located.

METHODS AND MATERIAL Hair follicles from the occipital scalp were obtained from 29 individuals.Measurements were performed on digital pictures using a software imaging system. Antibodyanticytokeratin (CK), 15 was used as a bulge stem cell marker.

RESULTS The mean length of a scalp hair follicle is 4.16 mm. The infundibulum measures 0.76 mm, theisthmus 0.89 mm, and the inferior portion 2.5 mm. The insertion of the arrector pili muscle is located1.65 mm deep. CK15 immunoreactivity starts at a depth of 1 mm and extends down to 1.8 mm.

CONCLUSION The ideal depth for the trichophytic procedure is to cut the wound edge at a depth of lessthan 1 mm to avoid the bulge zone. The data provided can serve as an objective anatomical reference inhair regeneration studies using horizontally transected follicles.

The authors have indicated no significant interest with commercial supporters.

The classical histological division of a human

hair follicle into three vertical portions (the

infundibulum, the isthmus, and the inferior portion)

serves as a frame of reference to delineate anatomic

compartments within the follicle. Briefly, the in-

fundibulum comprises the area of the follicle be-

tween the follicular ostia and the entrance of the

sebaceous duct, the isthmus is the portion between

the sebaceous duct and the arrector pili muscle, and

the inferior portion is the segment of the follicle

below the arrector pili muscle (Figure 1). The infe-

rior portion can also be divided into two portions:

the bulb and the suprabulbar segment.1 The entrance

of the sebaceous duct and the insertion point of the

arrector pili muscle are fixed structures, which

makes this simple classification useful and immuta-

ble throughout the hair follicle cycle. In contrast, the

inferior portion of the follicle undergoes an apopto-

tic regression in catagen and is absent in the telogen

phase of the hair cycle.

An anatomical landmark that has attracted

considerable attention for the past 2 decades because

of its role as at least one major site of follicular

epithelial stem cells is the bulge.2,3 The German

dermatologist Paul Unna first described the bulge, as

an anatomical term, as the portion of the hair follicle

to which the arrector pili muscle attaches4 (Figure 2).

The bulge can be easily identified in mice hair fol-

licles and in fetal hair follicles as a distinctive pro-

tuberance of the outer root sheath (ORS) at the

inferior portion of the isthmus. In contrast, this

protuberance is not morphologically detectable in

the majority of adult hair follicles, although ORS

& 2010 by the American Society for Dermatologic Surgery, Inc. � Published by Wiley Periodicals, Inc. �ISSN: 1076-0512 � Dermatol Surg 2011;37:58–64 � DOI: 10.1111/j.1524-4725.2010.01809.x

5 8

�Clınica Dr. Jimenez Acosta, Las Palmas de Gran Canaria, Spain; yRegenerative Medicine Program, InstitutoBiodonostia, San Sebastian, Spain; zDepartment of Pathology, Hospital General Universitario de Albacete, Universidadde Castilla La Mancha, Albacete, Spain

protrusions known as ‘‘follicular trochanters’’ can

occasionally be identified in the bulge region of

human terminal and vellus hairs.5,6

It is important to realize that, when we refer to the

bulge not in its anatomical sense but in its biological

sense (niche of follicular epithelial stem cells),

recent reports appear to indicate that its location in

human follicles extends beyond the limits of Unna’s

original anatomical description.3,7 The precise

location can be shown using immunohistochemical

biomarkers for human follicular stem cells that

reside in the bulge, such as cytokeratin 158,9 and

CD200.3,7 The immunoreactivity of these markers

shows that bulge stem cells may not be confined to

the arrector pili muscle attachment site, but rather

extend along the isthmus region.

Figure 2. On the left, a hematoxylin and eosin–stained ver-tical section of a scalp hair follicle showing the anatomicalbulge zone, described as the insertion site of the arrector pilimuscle (circle). On the right, the same section immuno-stained with the bulge stem cell marker CK15 showing thebiological bulge zone (brown). The length of the biologicalbulge stem cell region represents a wider zone (average 0.8mm), almost equivalent to the length of the isthmus (dis-tance from the sebaceous gland duct to the insertion of thearrector pili muscle).

Figure 1. A vertical section of a scalp hair follicle in anagenshowing the different portions of the hair follicle and thelengths that were evaluated in this study. L1 = distance fromthe skin surface to the entrance of the sebaceous duct,which correlates to the length of the infundibulum; L2 = dis-tance from the skin surface to the insertion of the arrectorpili muscle, which corresponds to the length of the in-fundibulum plus the isthmus; L3 = distance from the skinsurface to the base of the hair bulb, which corresponds tothe total length of the follicle. Asterisk shows the insertionsite of the arrector pili muscle. Ds, diameter superior (thick-ness of the hair follicle epithelium at the isthmus level); Di,diameter inferior (thickness of the hair follicle epithelium atthe inferior portion).

3 7 : 1 : J A N U A RY 2 0 1 1 5 9

J I M E N E Z E T A L

It would be interesting for hair transplant surgeons

and for hair regeneration studies (transplantation of

horizontally transected follicles) to understand as ac-

curately as possible at what depth below the skin

surface the different compartments of the hair follicles

are located, with special attention to the bulge stem

cell region. With this goal in mind, we undertook the

present histological and morphometric study.

Material and Methods

Anagen follicles were obtained from the occipital scalp

of 29 Caucasian volunteer individuals who underwent

hair transplant procedures. Seven patients were female

and 22 male, aged 26 to 67. From each individual,

three follicular units (each one containing from 1 to 4

hair follicles) were longitudinally sectioned in a cryo-

stat and stained with hematoxylin and eosin. Only

those whose vertical sections contained the complete

length of the follicle were measured.

The histological sections were visualized on the

computer screen and photographed with a digital

camera attached to an optical microscope (Motic

Group Co., Xiamen, China). Measurements of the

different portions of the follicle were performed on

the digital pictures using a software imaging system

(Motic Images Plus 2.0).

We measured the following segments of the follicles

(Figure 1):

(L1) = distance from the skin surface to the en-

trance of the sebaceous duct, which correlates to

the length of the infundibulum

(L2) = distance from the skin surface to the inser-

tion of the arrector pili muscle, which corresponds

to the length of the infundibulum plus the isthmus

(L3) = distance from the skin surface to the base of

the hair bulb, which corresponds to the total

length of the follicle

(Ds) = thickness of the hair follicle epithelium at

the isthmus level (excluding the dermal sheath)

(Di) = thickness of the hair follicle epithelium at the

inferior portion (excluding the dermal sheath)

Mean values and standard deviation were deter-

mined. From the above measurements, it can be

deduced that the isthmus corresponds to the L2

length minus the L1 length (L2–L1) and that the

inferior portion of the follicle corresponds to the L3

length minus the L2 length (L3–L2).

Immunohistochemistry

For immunohistochemical staining, we used human

scalp samples fixed in formalin. The samples were

deparaffinized and steamed for 40 minutes in citrate

buffer, pH 7, at 951C and immunostained with the

antibody anti-CK15 clone LHK 15 (Novocastra,

Newcastle, UK) using the standard avidin–

streptavidin staining method and the DakoCyto-

mation Autostainer (Dako, Glostrup, Denmark).

The incubation time with the primary antibody was

30 minutes.

The length of the CK15 immunoreactivity present

at the external root sheath of the follicle was mea-

sured in 20 terminal anagen hair follicles using an

ocular micrometer attached to the microscope.

The distance from the skin surface to the upper and

lower end of the CK15 immunoreactivity was also

measured.

Results

Hair Follicle Morphometry

Table 1 shows the results of the measurements of

each of the 29 patients included in the study and the

mean and the standard deviation. Table 2 shows a

summary of the most significant data. In short, the

mean length of a terminal anagen scalp hair follicle

(or the distance from the scalp epidermal surface

to the dermal papilla) is 4.270.4 mm. The

infundibulum measures 0.87 0.2 mm, the isthmus

(L2–L1) measures 0.9 mm, and the inferior portion

(L3–L2) is 2.5 mm long.

The insertion region of the arrector pili muscle to

the hair follicle, which coincides with the lowest

D E R M AT O L O G I C S U R G E RY6 0

M O R P H O M E T RY O F S C A L P H A I R

end of the isthmus, is located 1.67 0.3 mm from the

skin surface.

The mean diameter of the hair follicle (hair shaft

plus epithelial layers) is 1777 46mm at the level of

the isthmus and 254737mm at the inferior portion.

Length of the Bulge Zone (As Detected Using

CK15) and Depth Below the Skin Surface

CK15 immunoreactivity starts to be detected below

the entrance of the sebaceous duct into the follicle, at

an average depth of 1 mm (0.9–1.35) below the skin

surface and extends down to a depth of 1.8 mm (1.6–

2.25), just below the arrector pili muscle insertion.

The average length of the bulge region as detected

usng anti-CK15 is 0.8 mm, almost equivalent to the

length of the isthmus (Figure 2).

Discussion

In hair restoration surgery, knowing the length of the

terminal anagen hair follicle is important to deter-

mine the depth of the incision made at the recipient

TABLE 1. Measurements of the Different Portions of Occipital Scalp Hair Follicles in Caucasian Adults

Patient

mm

Length of

Infundibulum

Length of infundibulum

plus isthmus (from the

skin surface to the

arrector pili muscle

insertion)

Total

length of

the follicle

Diameter of hair

follicle (excluding

the dermal

sheath) at the

isthmus level

Diameter of

hair follicle

at the level

of the inferior

portion

1 1,184 2,127 4,584 100 265

2 1,038 2,097 4,829 116 211

3 774 1,527 4,741 124 298

4 1,043 2,238 4,289 127 241

5 702 1,460 4,114 129 266

6 773 1,340 4,661 131 326

7 460 1,322 4,366 138 256

8 666 1,367 3,809 142 266

9 888 1,882 4,620 144 205

10 933 1,955 3,803 150 284

11 852 1,850 3,630 164 192

12 728 1,445 4,150 165 275

13 879 2,002 3,786 168 266

14 996 1,983 4,645 169 273

15 629 1,292 4,087 169 243

16 475 1,204 3,664 172 274

17 633 1,459 3,520 172 271

18 718 1,712 3,592 183 279

19 879 1,917 4,040 186 207

20 678 1,472 4,628 190 221

21 734 2,022 4,556 192 169

22 769 1,561 4,324 200 274

23 991 1,936 4,024 206 297

24 438 1,261 3,931 207 274

25 652 1,599 3,649 224 229

26 553 1,450 4,366 247 269

27 831 1,678 4,413 259 195

28 564 1,300 4,562 264 250

29 624 1,432 3,165 283 287

Sample mean7standard

deviation

7627 187 1,6517 307 4,1567 439 1777 46 2547 37

3 7 : 1 : J A N U A RY 2 0 1 1 6 1

J I M E N E Z E T A L

sites. Several instruments can be used to create

recipient sites, such as chisel blades and needles. In

our study, the average length of the terminal hair

follicles obtained from a population of adults was as

4.16 mm, slightly longer than the data of Vogt and

colleagues (3.86 mm).10 This difference is not sig-

nificant, given that there is some interindividual

variability (the total length in our study group

ranged from 3.5 to 4.8 mm.) The follicular length of

approximately 4 mm concurs with the common

knowledge of hair transplant surgeons, which rec-

ommends creating recipient sites at a depth of 4 to

5 mm to accommodate the hair follicle graft,11 al-

though because of the interindividual variability and

significant ethnic differences in follicular size,12 it is

wise to adjust the depth of the recipient site on an

individual basis, matching it with the length of the

harvested follicle. The first author normally uses

custom-sized chisel-tipped blades to produce an ideal

match (length and width) for the follicular units and

a handle that allows blade length adjustment, giving

a limited-depth incision. Matching the size of the

recipient sites with the size of the follicular unit will

facilitate the process of graft insertion.

The depth of follicular section is important in the

so-called trichophytic donor closure hair transplant

technique.13 In this technique, one of the edges of the

donor wound is superficially cut, and the other edge

is overlapped on wound approximation and suturing

(Figure 3). The goal is for the transected hairs to

grow through the scar, with the aim of leaving an

almost undetectable scar. The depth at which the

edge is sectioned may be of relevance, because too

deep a cut may remove part or the entire bulge

region, resulting in permanent hair loss. The bulge

region is an important structure of the follicle

because it contains cells with stem cell characteristics

(multipotency, high in vitro proliferative potential,

and in vivo quiescence) important for the regenera-

tion of the pilosebaceous unit.3,14 Recent studies

have noted that the bulge stem cells are not only

located in the outer root sheath at the insertion site

of the arrector pili muscle (the anatomical definition

of the bulge), but also extend along the isthmus

portion of the follicle (Figure 2), although measure-

ments of the length and depth of this stem cell region

are lacking. Based on our study using the well-

known bulge stem cell marker CK15,8 the depth of

the bulge region extends from 1 mm from the skin

surface to 1.8 mm deep (total length of the bulge

0.8 mm), almost equivalent to the length of the

isthmus. Thus, on a practical level, the ideal depth to

perform the trichophytic procedure is to cut the

wound edge at a depth of less than 0.8 to 0.9 mm (at

the infundibulum level) to leave a safety margin. The

hair transplant surgeon should be especially careful

TABLE 2. Summary of the Most Relevant

Morphometric Data Obtained in This Study

Scalp Human Follicle

Mean

(mm)

Total length of the follicle 4.16

Infundibulum 0.76

Isthmus 0.89

Inferior portion 2.51

Diameter isthmus� 0.17

Diameter inferior portion� 0.25

Depth of the arrector pili muscle in-

sertion

1.6

Depth proximal end of the bulgey 1.0

Depth distal end of bulgey 1.8

Length of the bulgey 0.8

�The diameter at the isthmus level indicates the thickness of the

hair follicle excluding the dermal sheath.yBulge as defined by immunoreactivity to anti-CK15 antibody.

Figure 3. Cutting the inferior edge of the donor wound inthe trichophytic closure. Care must be taken to cut the edgesuperficially, above the sebaceous glands.

D E R M AT O L O G I C S U R G E RY6 2

M O R P H O M E T RY O F S C A L P H A I R

in avoiding any surgical transection or mechanical

trauma at the depth of the bulge zone (1–1.8 mm

from the skin surface). If the wound edge is cut

deeper than 1 mm, the transected follicle will lose

part of the bulge stem cell pool. It is unclear whether

leaving behind part of the stem cell pool would affect

hair growth, but care should be taken never to cut

the wound edge 1.8 mm deep or more, because at

this depth, the total bulge stem cell pool will be re-

moved, and the hair will probably not grow back.

The most evident clinical proof that the in-

fundibulum of the hair follicle can be removed

without affecting the regrowth of the hair is the scalp

wound created after harvesting split-thickness skin

grafts (STSGs). The scalp is an excellent source of

STSGs, not only because it has a healing capacity

faster than other regions, but also because the same

area can be used several times without permanent

hair loss.15–18 The reason is that these superficial

wounds are created by electrical dermatomes (Zim-

mer or Padgett) set at a thickness of 0.25 to 0.3 mm

so that the follicular epithelium removed is at the

infundibulum level, well above the depth of the bulge

stem cell region.

Another interesting observation of this study is that

an analysis of the proportions of the different hair

follicle compartments (infundibulum, isthmus, and

inferior portion) indicates that, of the total length of

the follicle (4.2 mm), roughly 60% corresponds to

the nonpermanent inferior portion (2.5 mm),

whereas 40% (1.6 mm) corresponds to the perma-

nent portion of the follicle (infundibulum plus isth-

mus). Therefore, if we transversely section a terminal

anagen hair follicle exactly in the middle, the supe-

rior half will always contain the bulge stem cell re-

gion and should theoretically have more regenerative

capacity than the inferior one (Figure 4). This is

important for hair regeneration studies using hori-

zontally transected follicles, because the level of hair

transection will determine the capacity of hair fol-

licle regeneration for each of the two portions.

According to Kim and Choi,19,20 when the follicles

are transected at the lower one-third, implants of the

upper two-thirds of the follicle produce a thick hair,

but implants of the lower one-third of the follicle do

not produce any hair, although a recently published

study by Toscani and colleagues21 had different re-

sults, showing that the implantation of the inferior

portion of the hair follicles (which does not contain

the bulge region) achieves a percentage of hair

regrowth (69%) similar to the implantation of the

superior (bulge-including) portion, supporting the

hypothesis of the presence of stem cells in the inferior

Figure 4. The photograph on the left shows one hair follicleunder the stereomicroscope. The ocular micrometer showsthat the length of this follicle is 4.4 mm. A horizontal line ismade at the middle of the follicular length (2.2 mm).The bulge stem cell zone should be located above thisline (within the oval shape between 1 and 2 mm). The panelon the right shows a histological vertical section of anotherhair follicle, immunostained with the bulge marker CK15.This panel shows that the bulge stem cell region is infact located approximately 1 to 2 mm from the skin surfaceusing the same magnification and ocular micrometer as forthe left panel.

3 7 : 1 : J A N U A RY 2 0 1 1 6 3

J I M E N E Z E T A L

portion of the follicle. Although more studies on this

fascinating area of hair follicle regeneration are

needed, we believe that the main aspect to consider

to achieve reproducible experiments is to strongly

encourage researchers to measure and document as

accurately as possible at what depth from the skin

surface the hair follicles are transected.

References

1. Sperling LC. Atlas of hair pathology. New York: Parthenon; 2003.

2. Cotsarelis G, Sun TT, Lavker RM. Label-retaining cells reside in

the bulge area of pilosebaceous unit: implications for follicular

stem cells, hair cycle, and skin carcinogenesis. Cell 1990;61:

1329–37.

3. Ohyama M, Terunuma A, Tock AL, et al. Characterization and

isolation of stem cell-enriched human hair follicle bulge cells.

J Clin Invest 2006;116:249–60.

4. Unna PG Beitrage zur Histologie und Entwicklungsgeschichte der

menschlichen Oberhaut und ihrer Anhangsgebilde. Arch Mi-

kroskop Anat Entwicklungsmechan 1876;12:665–741.

5. Narisawa Y, Khoda H. Two and three dimensional demonstra-

tions of morphological alterations of early anagen hair follicle

with special reference to the bulge area. Arch Dermatol Res

1996;288:98–102.

6. Tiede S, Kloepper JE, Whiting DA, Paus R. The ‘‘follicular tro-

chanter’’: an epithelial compartment of the human hair follicle

bulge region in need of further characterization. Br J Dermatol

2007;157:1013–6.

7. Inoue K, Aoi N, Sato T, et al. Differential expression of stem cell

associated markers in human hair follicle epithelial cells. Lab In-

vest 2009;89:844–56.

8. Lyle S, Christofidou-Solimidou M, Liu Y, et al. The C8/144B

monoclonal antibody recognizes cytokeratin 15 and defines the

location of human hair follicle stem cells. J Cell Sci

1998;111:3179–88.

9. Poblet E, Jimenez F, Godınez JM, et al. The immunohistochemical

expression of CD34 in human hair follicles: a comparative study

with bulge marker CK15. Clin Exp Dermatol 2006;31:807–12.

10. Vogt A, Hadam S, Heiderhoff M. Morphometry of human ter-

minal and vellus hair follicles. Exp Dermatol 2007;16:946–50.

11. Hasson V. Perpendicular angle grafting. In: Haber RS, Stough DB,

editors. Hair transplantation. Philadelphia: Elsevier Saunders;

2006. p. 35–42.

12. Mangelsdorf S, Otberg N, Maibach HI, et al. Ethnic variation in

vellus hair follicle size and distribution. Skin Pharmacol Physiol

2006;19:159–67.

13. Marzola M. Single-scar harvesting technique. In: Haber RS,

Stough DB, editors. Hair transplantation. Philadelphia: Elsevier

Saunders; 2006. p. 83–5.

14. Ohyama M. Advances in the study of stem-cell-enriched hair

follicle bulge cells: a review featuring characterization and isola-

tion of human bulge cells. Dermatology 2007;214:342–51.

15. Barret JP, Dziewulski P, Wolf S, et al. Outcome of scalp donor sites

in 450 consecutive pediatric burn patients. Plast Reconstr Surg

1999;103:1139–42.

16. Khalid K, Tarar MN, Mahmood F, et al. Scalp as a donor site for

split thickness skin grafts. J Ayub Med Coll Abbottabad

2008;20:66–9.

17. Mimoun M, Chaouat M, Picovski D, et al. The scalp is an

advantageous donor site for thin skin grafts: a report on 945

harvested samples. Plast Reconstr Surg 2006;118:369–73.

18. Weyandt GH, Bauer B, Berens N, et al. Split-skin grafting from

the scalp: the hidden advantage. Dermatol Surg 2009;35:1873–9.

19. Kim JC, Choi YC. Regrowth of grafted human scalp hair after

removal of the bulb. Derm Surg 1995;21:312–3.

20. Kim JC, Choi YC. Hair follicle regeneration after horizontal

resectioning: implications for hair transplantation. In: Stough DB,

Haber RS, editors. Hair transplantation: surgical and medical. St.

Louis, MO: Mosby; 1996. p. 358–63.

21. Toscani M, Rotolo S, Ceccarelli S, et al. Hair regeneration from

transected follicles in duplicative surgery: rate of success of cell

populations involved. Derm Surg 2009;35:1119–25.

Address correspondence and reprint requests to: FranciscoJimenez, MD, Calle Angel Guimera, 2, Las Palmas deGran Canaria 35003, Spain, or e-mail: fjimenez@clinicadelpelo.com

D E R M AT O L O G I C S U R G E RY6 4

M O R P H O M E T RY O F S C A L P H A I R