View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by Elsevier - Publisher Connector

Hair Transplantation: Current Concepts and Techniques

Walter P. Ungerw
Department of Dermatology, Mount Sinai School of Medicine, New York, New York, USA; wDepartment of Dermatology, Johns Hopkins School of Medicine,
Baltimore, Maryland, USA

Because of changes in technique, hair transplanting can now be offered as a reasonable option to more male and
female patients who are not responsive to, or likely to benefit from, medical treatment. These changes have also
resulted in exceptionally natural-looking results, even after a single session in an alopecic area, or in many
individuals with prior and cosmetically unacceptable transplanting results. Current concepts and techniques are
described in the article. Possible disadvantages of some of them are also discussed. In particular, the apparent
advantages of ‘‘megasessions’’ of 3000 or more grafts per session, ‘‘dense packing’’ of more than 40 follicular units
per cm2 and a new method of harvesting single follicular units directly from the donor area (Follicular Unit
Extraction), may not be advantageous as they first seem.
Key words: hair transplanting/current concepts/current techniques
J Investig Dermatol Symp Proc 10:225 –229, 2005

The basic building block of modern hair transplantation is (FPHL), who usually have less satisfactory long-term donor/
the intact ‘‘follicular unit’’ (FU). Table I summarizes the def- recipient area ratios than males, but who rarely lose all
inition of the different types of grafts used in transplantation of the hair in their thinning areas and who typically wish
today, based primarily on the number of intact FU they to have an appearance of maximum density. It is important
contain, and the type of instrument that is used to make to recognize that many women who fail to respond satis-
recipient sites for them (Unger and Beehner, 2004). Trans- factorily to medical treatment can be improved, and are
plant sessions that consist exclusively of FU (FUT) are ideal candidates for modern hair restoration surgery techniques
for transplantation of areas that are totally alopecic, for in- (Unger and Unger, 2003) (Fig 2).
dividuals who do not want or are not suitable for high hair The use of slot grafts or round grafts in portions of the
density objectives, and for those who are not ready to recipient area are indicated when hair density objectives are
commit to more than one session in any area. This is be- even higher, the long-term donor/recipient area ratio is
cause a single FUT session in an alopecic area—or an area judged to be good, and, in most cases, when there is a
that is destined to become alopecic—will appear com- moderate amount of original hair in the recipient area. Slot
pletely natural standing on its own (Fig 1) (Bernstein and grafts are particularly useful in Caucasian patients with
Rassman, 1997). Additional sessions are carried out in the white hair (Fig 3). In our hands, the ultimate in density ob-
same area only for increased hair density. jectives can be achieved with a combination of FU in the
If multiple-FU grafts (MUG) are utilized, they are used hairline zone, double FU (DFU) and triple FU (TFU) posterior
posterior to a hairline zone created exclusively with FU. In to the hairline zone, and round grafts posterior to the DFU
general, the larger the MUG used (the more FU in the graft), and TFU. Posterior to the round grafts, DFU and TFU are
the greater the density that can ultimately be created, and continued. Ideal candidates for this type of grafting have the
more importantly, the safer the FU are within the graft from following characteristics: (a) a good long-term donor/recip-
potentially lethal technical mistakes that can be made during ient area ratio: (b) good long-term temporal hair density: (c)
graft preparation, storage, and insertion (Unger, 2004a). some hair in the recipient area: and (d) good hair charac-
Once MUG are used, however, with a few exceptions, a sin- teristics for hair transplanting. A large majority of patients do
gle session in an area that is alopecic or destined to become not qualify based on the above factors. Therefore, this op-
so will not appear sufficiently natural to ‘‘stand on its own.’’ tion is only offered to approximately 5%–10% of the pa-
Combining a hairline created with FU and micro-slit tients that the author sees.
grafts posterior to that is most suitable for patients who (a) The decision as to what type of graft to use in what cir-
want higher ultimate hair density than can be accomplished cumstances is far too complicated to adequately deal with in
with FUT if one utilizes reasonable FU densities per session; the space allotted for this discussion. The reader is advised
(b) have persisting hair in the recipient area; and (c) have to review the 2004 edition of Hair Transplantation (Unger and
appropriate long-term donor/recipient area ratios. One ex- Shapiro, 2004) for more complete information. It is worth-
ception to the latter is women with female pattern hair loss while, however, to note here that the noticeability of a graft is
not just a function of its size. The less the contrast between
hair and skin color, the finer the hair texture; the frizzier or
Abbreviations: FU, follicular unit; DFU, double FU; TFU, triple FU curlier the hair, the larger the graft can be without it being

Copyright r 2005 by The Society for Investigative Dermatology, Inc.

225
226 UNGER JID SYMPOSIUM PROCEEDINGS

Table I. Definition of the different types of grafts used in transplanting

Cut ‘‘to size’’aor cut to ‘‘number

Graft type Hairs FU Recipient site of hairs’’b
Micrografts
Micrograft (general term) 1–4 1 or less Needle/micro-slit Cut to ‘‘number of hairs’’
FU) (specific term) 1–4 1 Needle/micro-slit Cut to ‘‘number of hairs’’
FF 5–6 2 Needle/micro-slit Cut to ‘‘number of hairs’’
Multi-FU graft
Micro-slit grafts
DFU 3–5 2 Slit Cut to ‘‘number of hairs’’ (FU)
TFU 5–8 3 Slit Cut to ‘‘number of hairs’’ (FU)
QFU 6–12 4 Slit Cut to ‘‘number of hairs’’ (FU)
Traditional slit grafts

Small slit graft 3–5
2 Slit Cut to size

Medium slit graft 5–8
3 Slit Cut to size

Large slit graft 6–12
4 Slit Cut to size
Slot grafts
Small slot graft 6–8
4 Slot Cut to size
Medium slot graft 8–12
6 Slot Cut to size
Large slot graft 10–16
8 Slot Cut to size
Round grafts
Small round graft 5–8 2–3 Punch Cut to size
Medium round graft 8–14 4–5 Punch Cut to size
Large round graft (includes 14–30 þ 6–15 þ Punch Cut to size
‘‘standard’’ punch grafts)

FU, follicular unit; FF, follicular family; DFU, double FU; TFU, triple FU; QFU, quadruple FU.
a
‘‘Cut to size’’ indicates that the number of hairs in the graft is less important than its size and how well the graft fits into the size of the proposed
recipient site.
b
Cut to ‘‘number of hairs’’ denotes that the primary concern of the graft cutter is the number of hairs in the graft rather than its size.

readily noticed. As implied earlier, often most importantly, the serve of what is expected to be a permanent hair-bearing
more the persisting hair in the recipient area, the larger the donor area should be maintained for as long as possible. In
graft can be without it being noticeable (Unger, 2004a). addition, in males, the lateral borders of the transplanted
‘‘Repairs’’ of esthetically unsatisfactory ‘‘old’’ hair trans- area should be constructed with FU and/or DFU. Should
planting are sometimes treated exclusively with FU, or a one misjudge the ultimate limits of the MPB, the patient
combination of FU with excisions of part or all of some will thereby be left with a larger than usual, but sometimes
grafts—or completely untreated alopecic areas. But many naturally occurring, soft-bordered isolated frontal forelock
of these patients are best treated with combinations of (IFF). The latter, because it occurs naturally in some indi-
these techniques and larger MUG (Unger, 2004b). viduals, will appear natural standing on its own (Unger and
Beehner, 2004).
Conceptually, we think of the recipient area as consisting
Planning in Hair Transplantation of three areas: a frontal, mid-scalp, and vertex area. Typ-
ically, we treat only one of those areas during any given
Male pattern baldness (MPB) and FPHL are progressive session. Transplanting the frontal, as well as mid-scalp ar-
disorders. Planning hair transplantation should therefore eas—in two or more sessions—as far posteriorly as the
ideally include an allowance for future areas of involvement, point where the scalp changes from being more or less
by extending the grafting into areas that are still hair bearing parallel to the ground to more or less perpendicular to it
but that can be reasonably expected to lose hair in the results in the individual appearing to have hair from both
future (Fig 1). Such an approach potentially avoids a con- frontal and lateral views. This should be a minimum objec-
stant and frustrating ‘‘chasing’’ of an enlarging alopecic area tive in most patients. A trial of medical treatment for MPB in
(Unger and Beehner, 2004). the mid-scalp and vertex areas should also be encouraged
Because perfect prognostication of the degree of alo- before a surgical procedure is carried out, especially if
pecia that will develop in any patient is impossible, a re- there is persisting hair in these areas (Knudsen, 2004). The
10 : 3 DECEMBER 2005 HAIR TRANSPLANTATION 227

Figure 1 Figure 2
‘‘Before’’ and ‘‘After’’ FUT. ‘‘Before’’ and ‘‘After’’ FU/Micro-slit grafting.
(a) Photograph of a patient before his first hair transplant procedure. (a) A female patient before transplant. The black crayon marks delin-
The black crayon marks delineate the proposed areas for treatment. eate the area to be treated. (b) Nine months after a single session
Note also the inclusion of lateral areas where some hair persists but can in which follicular units (FU) were used in the hairline zone and double
be anticipated to be lost in the future; these areas will be transplanted FU (DFU) were used posterior to the hairline zone. The surgical
at the same time as the more obvious areas of alopecia are being treatment of female pattern hair loss, in appropriately selected in-
treated. (b) This photo was taken 12 mo after the second session of dividuals, is a worthwhile treatment to be kept in mind if medical treat-
follicular unit transplanting. The first session was carried out in the ment fails.
frontal area and the second was carried out in the mid-scalp area.

vertex area, when transplanted, is nearly always treated nation of traction and peripheral 27-G needle punctures
only with FU, primarily because of limited donor reserves in intended to sever its fibrous attachments. It is an advan-
most individuals. tageous technique for special circumstances, such as in
patients who are prone to develop wider than normal scars
or who have such scars and for patients with tighter than
The Donor Area average scalps. The concept, unfortunately, currently has
several serious problems associated with it and therefore—
In most patients, the first donor strip is excised from the despite extensive Internet hype—cannot be recommended
densest areas of the permanent zone—the areas that are by the author for routine use in most individuals (Unger and
also least likely to be affected by future hair loss (Unger, Cole, 2004).
2004c). It is approximately 8–10 mm wide, and begins su-
perior to the left ear and ends superior to the right ear.
Donor strips for all subsequent sessions are taken such that Graft Preparation and Placement
the scar from any preceding session(s) is included in the
center of the new strip. Thus, no matter how many sessions One of the largest changes in the evolution to modern
are carried out, only a single scar is ever produced. hair transplantation has been the large number of grafts
A recent and sometimes useful innovation in harvesting that are characteristically transferred during each session.
donor tissue has been referred to as FU Extraction This has resulted in substantially more hair being
(FUE) (Rassman et al, 2002). A small bore trephine— transplanted per session—and other important advan-
usually 1 mm or less in diameter—is used to incise around tages—but has also introduced a much greater potential
an individual FU to an approximately mid-dermis level. problem of lethal follicular damage during the production
The FU is then gently eased out of its site with a combi- and/or storage and/or placement phase of surgery.
228 UNGER JID SYMPOSIUM PROCEEDINGS

grafts, or 2 FU wide for slot grafts, or three or more FU wide

for ‘‘round’’ grafts of various sizes. A stereomicroscope
(  6 magnification) is utilized for the ‘‘slivering.’’ Sub-sec-
tioning of the ‘‘slivers’’ can be performed using the stereo-
microscope or lesser magnification, depending on the
visual acuity of the technician.

The Recipient Area

FUT Whenever an area is being treated exclusively with FU,
we would generally use a density of approximately 20–25
FU per cm2 per session. Studies showing different rates of
hair survival with different FU densities have indicated that
when one goes beyond these FU densities, the survival
rates tend to drop to what we feel are unacceptable levels
(Mayer and Keene, 2004). The results of the second of two
studies are summarized in Table II. Recipient sites are gen-
erally made with an 18- or 19-G needle with its sharp end
bent so as to limit the depth to which it will penetrate the
skin or small custom-made blades. As with all types of
grafts, incisions are made at the same angle and direction
as the original hair in that area.
Within the recipient area, the FU may be transplanted in
somewhat greater or lesser densities per cm2 and they are
chosen specifically for the caliber and number of hairs in
each of them (Shapiro, 2004). Two FU that are close enough
together to fit into a single needle hole may be left as an
intact unit during graft preparation in order to increase the
number of grafts with four or more hairs and thereby in-
Figure 3 crease the hair density in specific sites. Two individual FU
‘‘Before’’ and ‘‘After’’ FU/Micro-slit/Slot grafting. may be placed in a single needle hole for the same purpose.
(a) Patient before hair transplant. (b) The same patient 1 y after two The latter two types of grafts are, respectively, referred to
sessions of hair transplanting consisting of follicular units (FU) to the
hairline zone, double FU (DFU), and triple FU (TFU) posterior to the FU as ‘‘follicular families’’ (FF) and ‘‘paired’’ FU (Harris, 2004;
area and slot grafts posterior to the zone treated with DFU and TFU. Shapiro, 2004).
This combination of grafts is particularly advantageous for Caucasians The author believes that the recent trend of transplanting
with white hair, as substantially better density can be produced with it
than is possible with follicular unit transplanting (FUT) or a combination
3000 or more FU per session, by some practitioners, should
of FU and micro-slit grafts. be viewed with some skepticism with regard to hair survival.
For example, although each 18-G needle hole may appear
Space does not allow us to go into all the details of small, and typically produces an incision that is ‘‘only’’ ap-
graft preparation, storage, and placement in this chapter, proximately 1.2 mm long, 3000 of these incisions will pro-
so the reader is referred elsewhere for this information duce a total of 360 cm (12 ft) of incisions in an area that is
(Rose and Shapiro, 2004). There are several aspects of often no larger than the palm of an average man’s palm.
current techniques that the author believes should be Intuitively, such an approach, as well as FU densities of 30,
mentioned here: 40, 50, or more per cm2 (respectively 3.6, 4.8, and 6.0 cm of
Once the strip has been removed from the donor area, it incisions in each cm2 of recipient area) should sound alarm
is immediately placed in a chilled saline solution. It is then bells until satisfactory hair survival studies are produced by
sectioned—much as one would slice a loaf of bread—into the practitioners who advocate these approaches. Compli-
‘‘slivers’’ of tissue that are 1 FU wide for FU and micro-slit cating matters is that the advantages of high FU per cm2
densities and large numbers of FU per session are obvious,
Table II. FU survival: Mayer and Keene whereas their probable cost in hair survival is invisible. For
example, if 50 FU per cm2 results in an 82% FU survival
Number of 2-haired Number of new Percent rate, as suggested by Mayer and Keene’s study (Mayer and
FU inserted FU at 9 mo survival Keene, 2004) approximately 40 FU per cm2 will survive and
20 19 95 grow. The results of 50 FU per cm2 will appear to be more
30 23 76.7 than twice as good as 20 FU per cm2 (with 95% hair sur-
vival) but at the possible cost of a 15% higher rate of death
40 28 70
of irreplaceable and limited donor hair.
50 41 82
Micro-slit grafting Sites for DFU, TFU, and quadruple
FU, follicular unit.
FU are made 2–3 mm apart laterally and 1–2 mm apart
10 : 3 DECEMBER 2005 HAIR TRANSPLANTATION 229

antero-posteriorly (Unger, 2004d). After a single session, the planting is an option for their male and female patients who
total number of FU in the recipient area might be quite have not responded satisfactorily to medical treatment, and
similar whether FUT or micro-slit grafting is used—and in for those who had cosmetically unsatisfactory earlier types
most cases, FUT will look better. But the advantage of the of hair transplantation.
mixture of grafts is that a second or third session of micro-
slit grafting in the same area can easily double or triple the
density in that area, whereas this is not possible with FUT at DOI: 10.1111/j.1087-0024.2005.10111.x
acceptable FU densities.
Manuscript received September 20, 2004; revised November 2, 2005;
accepted for publication March 9, 2005
Slot grafting Slot grafts sites are spaced approximately 2–
3 mm lateral to each other and 1–2 mm apart antero-pos- Address correspondence to: Walter P. Unger, MD, FRCP(C), FACP, 99
teriorly, and are usually prepared in a pattern that is three or Yorkville Ave., Ste. 214, Toronto, ON M5R 3K5. Email: [email protected]
four rows deep (Unger, 2004d). Specially, designed ‘‘slot
punches’’ are utilized to prepare the recipient sites (Fig 3).
References
‘‘Round’’ grafting As noted earlier, in a relatively small
percentage of our patients, a hairline will be created with Bernstein WR, Rassman WR: Follicular transplantation: Patient evaluation and
surgical planning. Dermatol Surg 23:771–784, 1997
micrografts followed by micro-slit grafts more posteriorly Harris J: Recombinant follicular units: Concept formalization. In: Unger WP,
and then a zone of ‘‘round’’ grafts approximately three to Shapiro R (eds). Hair Transplantation, 4th edn. New York: Marcel Dekker,
four rows deep. The round recipient sites are, in fact, oval 2004; p 469–475
shaped because they are cut out at an angle and direction Knudsen R: Effect of medical therapy on surgical planning. In: Unger WP, Shapiro
R (eds). Hair Transplantation, 4th edn. New York: Marcel Dekker, 2004; p
that mimics that of the existing hair and, as implied earlier, 146–151
the grafts are more square than round. Each site is created Mayer M, Keene S: Mayer’s and Keene’s study comparing FU growth with dif-
approximately one-punch diameter width apart from its ferent planting densities. In: Unger WP, Shapiro R (eds). Hair Transplan-
tation, 4th edn. New York: Marcel Dekker, 2004; 921pp
neighbor in all directions (Unger, 2004d). It is important to
Rassman WR, Bernstein R, McClellan R, Jones R, Worton E, Uyttendaele H:
keep in mind that whenever slot grafts or round grafts are Follicular unit extraction: Minimally invasive surgery for hair transplanta-
utilized, patients are warned to return for a second session tion. Dermatol Surg 28:698–703, 2002
in approximately 5 mo, or the grafts may become notice- Rose P, Shapiro R: Combining microscopic slivering with backlighting and loupe
magnification to efficiently produce grafts. In: Unger WP, Shapiro R (eds).
able. After two sessions of either slot grafts or round grafts Hair Transplantation, 4th edn. New York: Marcel Dekker, 2004; p 363–372
in the same area, hair density is usually sufficient such that Shapiro R: Follicular unit transplantation alone or follicular units with multi-FU
simply running your fingers through the hair makes any graft grafts: Why, when, and how? In: Unger WP, Shapiro R (eds). Hair Trans-
plantation, 4th edn. New York: Marcel Dekker, 2004; p 435–469
noticeability (for example, when the hair is wet and parted
Unger W, Beehner M: Planning and organization of the recipient area. In: Unger
through the transplanted area) disappear. Sometimes, a WP, Shapiro R (eds). Hair Transplantation, 4th edn. New York: Marcel
third session is carried out in a slot-grafted area, whereas a Dekker, 2004; p 81–146
third session is virtually always carried out in areas treated Unger WP: Why ‘‘mixed’’ grafting. In: Unger WP, Shapiro R (eds). Hair Trans-
plantation, 4th edn. New York: Marcel Dekker, 2004a; p 663–687
with round grafts. A minimum of 5–6 mo is recommended Unger WP: Correction of cosmetic problems in hair transplanting. In: Unger WP,
between sessions if they are carried out in the same area. Shapiro R (eds). Hair Transplantation, 4th edn.. New York: Marcel Dekker,
2004b; p 663–687
Unger WP: The Unger approach to the donor area. In: Unger WP, Shapiro R (eds).
Conclusion Hair Transplantation, 4th edn. New York: Marcel Dekker, 2004c; p 337–
342
Unger WP: How I use multi-FU grafts. In: Unger WP, Shapiro R (eds). Hair Trans-
Modern techniques in hair transplantation have resulted in plantation, 4th edn. New York: Marcel Dekker, 2004d; p 503–516
far more patients being eligible candidates for this type of Unger WP, Cole J: Follicular unit extraction. In: Unger WP, Shapiro R (eds). Hair
surgery and the production of remarkably natural-looking Transplantation, 4th edn. New York: Marcel Dekker, 2004; p 334–337
Unger WP, Shapiro R (eds). Hair Transplantation. New York: Marcel Dekker,
results. Unfortunately, not all recent innovations are without
2004
likely or acceptable ‘‘costs’’ that may or may not be obvi- Unger WP, Unger R: Hair transplanting: An important but often forgotten treat-
ous. Physicians are urged to keep in mind that hair trans- ment for female pattern hair loss. J Am Acad Dermatol 49:853–860, 2003