Good News and Bad News in Hair Restoration Surgery (October 2011)
Hair Transplanting in Early Stage Male Pattern Baldness (September/October 2011)
Candidacy of females for hair transplantation (July/August 2011)
For more publications, please read Dr. Unger’s Curriculum Vitae…
The Principles of Good Long-Term Planning in Hair Transplantation
I appreciate being asked by Brian to write an essay for hairlossexperiences.com. Despite having been previously asked to write chapters on hair transplanting for 37 dermatologic and plastic surgical textbooks over the last few decades (the last three since 2012) and editing or co-editing five multi-authored textbooks entirely devoted to that subject (the last one in 2011), I have essentially not written for the general public, so the invitation was a welcomed one. Until now, I have always assumed that by writing for physicians, the contents of those publications would indirectly reach and help the largest number of people who aren’t physicians. The world has changed however and perhaps this presentation will actually reach more people than any of my more recent professional publications. We shall see.
This communication is intended to offer what I think is good general long-term planning for hair transplanting—and is directed at those looking for that sort of advice. If you are not as interested in what might happen to your current hair transplant goals over the next one or more decades, as you are about what you want now, you should stop reading this. However, most of you who think that way are destined to be very sorry in the long run unless new and better drugs are developed that stop or reverse male pattern baldness (MPB) and female pattern hair loss (FPHL) or unless successful cell therapy (what the general public often call “cloning”) is at last available, before you run into the consequences of your earlier choices. Such short-term choices, of course may be worthwhile for a minority of individuals whose present appearance is more important to them than probable long-term consequences (for example, entertainers and models) but not for most of you.
The photos I use in the discussion are intended to validate my points of views. Some of you may not believe this, but it is important for me to emphasize that I am not looking for surgical patients. I have been fortunate enough to continue to have more than enough patients without belonging to this or any other “expert” sites and many of you may not even have heard of me. I also don’t reply to emails from people I have never actually met in a prior in-person consultation. It’s not that I think I am too important to answer questions, it’s that I am too busy operating and consulting with potential patients while trying to give my wife and family more of my time than I have for many years. (Incidentally, patients pay a fee for my consultations.) I have answered many of the questions you may have, on my personal website (www.drwalterunger.com) so you could look there for answers to at least some of your questions and also easily access the more recent references I refer to in this submission via my website. Good luck to all of you.
Let’s begin by recognizing the realities of MPB and FPHL:
a) Both MPB and FPHL will progress intermittently but inevitably for as long as you live. There is no age, even advanced age, at which you can confidently predict that things will get no worse. Put more plainly, the areas of partial to complete hair loss will periodically get larger and the intervals of stability and additional hair loss will continue to occur after variable periods of time from person to person. There is no way to know how long these periods will be for you or anyone else.
b) In pace with (a), the potential donor area (the fringe of persisting hair in someone with MPB) will get smaller. Unfortunately this means the larger the bald areas are destined to become, the less hair you will have to supply the donor hair that you need to treat these enlarging areas. Those of you who have family histories that include individuals with more severe forms of MPB and FPHL – for example, Hamilton–Norwood Types VI or VII or Ludwig Type III (Figures 1 & 2) should be especially cautious about using up large amounts of donor hair at the time of your earliest occurring problem areas. This is especially wise if you are in your 20s or early 30s, because using very large numbers of hairs/grafts early on when the ultimate extent of your hair loss is less predictable, increases the chance that in the long run, you may eventually end up having large thinning or bald areas for which too few or no hairs/grafts will be available to adequately treat them. This means, for example, that while spectacularly dense and youthfully low hairlines can be created by using “dense packing” of 40, 50, 60 or more “follicular units” (FU)/cm2, such choices are extraordinarily unwise for nearly all young men who, on careful examination, seem likely to develop or who have family histories that include individuals who eventually developed the more severe forms of MPB (especially Types VI and VII).
The patient shown in Figure 3 was one of those who had such a family history. He had originally wanted transplanting up to the most forward of the two frontal hairlines I had outlined in the “before” photos. He had seen websites in which that sort of objective was presented as being a reasonable one for young men fighting their receding hairline. The higher hairline was the one I advised him to accept. The area between the more frontal drawn hairline and the higher one that I had recommended would have required “dense packing” of approximately 4000 FU according to the owners of a few websites that this patient consulted. After examining him, there was little doubt in my mind that he would eventually (and probably soon) evolve to a Type VI or worse MPB. As noted in Table 1, which will be discussed immediately below,(1) it is likely that this man will ultimately, on average, only have approximately 5393 FU containing likely permanent hair available over his lifetime. If he had gone to one of the doctors whose sites offered him the “dense packing” and lower hairline, he would have ended up with a very dense and low frontal hairline zone which he would have been very happy with for the time being. But eventually he would have had an extraordinarily large area of baldness behind that, and a grossly unnatural appearance. Unfortunately, a few doctors with very active Internet sites suggest the lower hairline or one similar to it, as well as very high hair density goals because this is what the patient wants rather than what he should want.
As alluded to above, this begs a question: “How many FU containing very likely permanent hairs are available anyway?” Obviously, the answer will depend on how severe your MPB is likely to become, as well as how dense your fringe hair is and is likely to become as you get older. Just as clearly, there is no one answer for everybody, but in order to get some approximate idea of the answer to that all-important question, I asked the following questions to the members of an “invitation only” small email group of very experienced and well-recognized world experts in hair restoration surgery, to which I belong:
“Keeping in mind that over the years, the hairs closest to the upper, lower and frontal borders of the fringe will be lost, how many FU containing very likely permanent hairs can be harvested from:
(a) a 30-year-old patient who you believe is destined to develop Type V MPB and has:
(1) higher than average hair density;
(2) average hair density;
(3) less than average hair density
(b) the same question but for a patient you believe is destined to evolve to Type VI MPB.”
A summary of the answers of 39 of us (including myself) are shown in Table 1. (See addendum for their names.) These numbers—while they are not dogmatically applicable to all individuals—are far from meaningless given their source and would be worthwhile keeping in mind by anybody carrying out or undergoing hair restoration surgery, so I strongly recommend that you also read the paper from which it is excerpted. As noted earlier, it can be accessed through my website (www.drwalterunger.com) and includes clarifying comments from some of the responders. Notwithstanding the lack of scientific validity of the tabulated answers, they are based on over 900 years of the combined experiences of many experts, and in my judgment, if the readers remember nothing else in this correspondence, they should remember the numbers in Table 1.
Let’s digress for a moment as to why I used bold lettering for “containing very likely permanent hairs” in my question:
First, as most of you probably already know, transplanted hair only lasts as long in the area into which it was placed, as it would have in the area from which it was taken. THE priority in hair transplanting has therefore always been to try to use only those grafts/hairs that are most likely to be in “permanent” hair-bearing areas. (The quote marks are there because nobody can be absolutely certain which areas will actually be permanently hair-bearing as opposed to most likely to be permanent in any particular individual.) In addition, not only does the fringe hair in a person with MPB get narrower and narrower with the passage of time, but the hairs within it also become finer-textured, and some of them—especially around the periphery of the fringe—are eventually lost.
In the early days of transplanting, many doctors and their patients got into a great deal of trouble because of the above phenomena. We didn’t realize how severely and quickly the fringe hair might narrow, and even less how fine the hair in the persisting fringe would become, because no studies had ever been done on those components of MPB and FPHL. (Finer caliber hairs produce less coverage/apparent density than comparable numbers of hairs with larger calibers. For example, Cole has noted that increasing the diameter of an average hair by just 0.01mm increases its coverage by 36%!) (2) As a result, some transplanted hairs were eventually lost and donor scars that weren’t initially noticeable in the harvested fringe areas became noticeable, at first only when the hair was wet and then later even when the hair was dry (Figure 4). Where hair was totally lost in the prior donor areas, the scars had no camouflage at all!! Of course, those hairs were also lost in the recipient areas into which they had been transferred. The consequences of the preceding phenomena were two of the worst long-term complications that could be imagined by physicians and patients, and still haunt us.
When I started seeing those complications in some of my patients and those of my colleagues, I decided to try to establish the most likely boundaries of the fringe area that might contain the largest number of hairs that were the most likely to be permanent. I, and a resident physician I was training, carefully examined the fringe hair of 328 men 65 years old or more, looking for areas where there persisted at least 8 hairs/4mm diameter circle. Using the data we had collected, we created those boundaries and called it the “Safe Donor Area” (SDA). I published the information (3) and it remains a widely accepted guide to ethical practitioners, as to where they should ideally or “most safely” obtain their grafts for transplanting. (The word “guide” is in bold letters to designate that it is only a guide or warning, rather than a dogmatic necessity to stay within the SDA. Clinicians should also use clinical examination, the patient’s age, family history, etc., to decide if the boundaries can be reasonably enlarged or reduced for any given individual.) Inherent in the preceding, it would seem to be safest for hair longevity to harvest grafts within the densest fringe hair for all sessions, staying as far away from frontal, upper and lower fringe borders—from which hair loss progresses. That has been best accomplished by first harvesting a strip of hair-bearing “donor” tissue from the densest hair, and in subsequent sessions including the scar from the prior one(s) in the new strip. The physician, with each harvest, thereby obtains not only the next most likely permanent hairs but also leaves behind only one scar running through the densest fringe hair, regardless of the number of sessions the patient has undergone.
It has been estimated that the hairline zone is approximately 30 cm/2, the frontal-third of a patient with Types V to VI MPB is approximately 70 cm/2, the frontal-half of the area of MPB is approximately 100 cm/2, the frontal two-thirds approximately 130 cm/2, and the full area of MPB approximately 230 cm/2.(4) (Other practitioners have estimated larger total areas of alopecia in Type VI MPB.) Yet, even if the frontal-half of an area of MPB is only 100 cm/2 and it were proposed to treat that area with an average density of 60 FU/cm2, 6000 grafts would be required to accomplish that. Reviewing Table 1, one can see that virtually all of the grafts in a patient who was destined to develop Type V MPB would have been used to treat only a small portion of the eventually bald area. If he were destined to develop Type VI MPB, not even the frontal-half of the area of MPB could be transplanted. This outcome can be partially avoided by doing the highest FU densities in the most cosmetically important areas, for example the hairline zone, and lower densities further back. But this is a very unusual hair dispersion—frontal scalp hair is typically least dense in the hairline zone and becomes progressively denser farther back. Also, even at 30 FU/cm2 I typically use approximately 1200 FU (or more) for a hairline zone that is approximately 2.5cm (one inch) deep so an FU density of 60 FU would use 2400 FU and 90 FU would consume 3600 FU in that narrow zone alone. Yet, as will be discussed below, on some Internet sites, “dense packing” of 60, 80, 100 and more FU/cm2 continues to be suggested as one of the best options for young men who want to replace their teenage hairlines.
c) Figure 5 consists of photos of a young medical student who I recently saw and treated, using Follicular Unit Extraction (FUE) to improve a FUE procedure that had been done by another hair restoration surgeon. The patient had chosen FUE because he had a family history of severe MPB accompanied by relatively low fringe hair density. He wanted to be able to keep all of his scalp hair—recipient and donor area fringe—very short without any scarring being noticeable and hence, chose FUE instead of strip harvesting—which, as noted earlier, always leaves at least one linear scar. The physician who operated on him however, took the grafts too close together. He did that because he was trying to maximize his hair graft yield, in order to maximize the areas he could transplant. He was also thinking long-term and trying not to harvest hair from areas above or below the SDA, where more hairs were more likely to be lost with the passage of time. Unfortunately, the result of the above seemingly reasonable two goals was that when the patient tried to wear his hair as short as he had planned, the donor areas looked somewhat “moth-eaten” (Figure 5a). FUE has some important and widely known advantages for some individuals but it is not scarless and without its own potentially serious problems. I tried to correct the problem by a) taking grafts via FUE that were as close as possible to the previously harvested areas (so the most likely to persist for his lifetime) and b) putting them into the worst scars from the prior harvest. If the first surgeon and the patient had been satisfied with fewer grafts this problem could have been avoided. If they had instead spaced the extractions farther apart they would have needed to go higher and/or lower to get an equivalent number of grafts. But those higher and lower grafts would have contained hairs that were increasingly more likely to not be permanent the higher and lower they were obtained! (Figure 4) This is the unavoidable dilemma of FUE. It is also why I recommend that most younger patients not use FUE— because the size of their ultimate donor area is less predictable than in older individuals.
Moreover, while this patient had more or less immediate problems with scar noticeability in his donor area, many patients undergoing FUE, who do not initially have similar problems, will eventually develop them as their donor site hair density and hair caliber decreases over time. The more extensive the harvesting area and extraction concentration, the more likely that will occur. FUE hasn’t been around long enough for us to know how severe and frequent the problems will become but these small grafts are being harvested in the same way as the old round grafts of early transplanting were (punch extraction) and too often over an “old fashioned” similarly large portion of the fringe hair. As discussed earlier, those early scars became increasingly visible as the hair around them was lost (a) partially or (b) completely. Please note: The above discussion on potential FUE disadvantages, applies to all FUE techniques, regardless of whether the FU are extracted manually or by a motorized device such as Neograft or by a robotic device such as the ARTAS® robot (for which I was a medical consultant in its development from 2003 to approximately 2012 and an owner of 100,000 shares of the company).
d) “Dense packing” of grafts/hairs to recreate the hair density you once had—especially if the loss was relatively recent—is very understandable. But before you think you need dense packing to look “natural”, you should also be introduced to a study on hair density done by Dr. Sharon Keene on “normal men”, none of whom had a family history of MPB or themselves had any noticeable MPB.(5) It contradicts a core belief of physicians who have thought for many years that “normal” scalp hair density was 100 FU/cm2—because of a study published in 1984.(6) What follows is an excerpt from Keene’s contribution to the fifth edition (2011) of Hair Transplantation – the reference textbook I co-edit: (All bold print and parentheses words are my translations of technical words or my emphasis.) Hold your breath because it is a long excerpt but a VERY important one:
“The author defines “cosmetic” density in hair transplantation surgery as a combination of natural hair distribution, good volume, and blocking of light from the scalp—or lack of visual evidence of hair loss regardless of hair or graft numbers. In contrast, “maximum” density refers to transplanting follicular unit (FU) grafts at high numbers per square centimeter (>50 FU/cm2) without particular consideration to hair numbers. While the goal of achieving maximum density sounds attractive to any patient with hair loss, the reality is that all hair restoration procedures on patients with Class IV or greater hair loss must work with areas of hair loss that are greater in size than areas of permanent donor hair that are available to cover them truly densely. Furthermore, the patient’s age and family history must be factored into any hair redistribution plans since a patient with, for example, a Class II or III Hamilton–Norwood pattern at presentation could eventually progress to a Class IV or greater pattern in the future. And, finally, although case reports of excellent hair survival at high FU density can occur in a single square centimeter, surveys suggest these densities are neither necessary nor naturally occurring (see below). Placing more hairs than what is required to achieve natural esthetic results, is therefore not advantageous to a patient, and in fact most often represents excessive and unnecessary use of a finite donor resource (as well as an unnecessary expense). Prudence would dictate that except for patients with stable patterns (if there is such an entity) and minimal hair loss, transplanting the least number of donor hairs necessary to achieve natural-appearing cosmetic density is of maximum advantage to patients with hair loss compared to achieving a goal of maximum density of grafts.
In our own pilot survey of naturally occurring hairline density, in 14 males with shaved heads and no personal or family history of androgenetic alopecia, (emphasis by WU) findings supported previous anecdotal observations in this practice. Temporal (“temple”) point density ranged from 24 to 59 FU/cm2 (average 41) and anterior (“frontal”) hairline density ranged from 38 to 78 FU/cm2 (average 52) (1). It was also noted that occipital (“back of head”) donor hair density largely mirrored the findings of Jiminez et al., who reported a range of 65 to 85 FU/cm2.(7) In addition, when the three-to-four hair FU were more numerous, FU density decreased but hair counts per square centimeter were roughly comparable. The observed FU hair densities were also consistent with the findings of Rassman and Bernstein, who reported an average of 2.2 hairs/FU.(8) It is worthwhile noting that such hair densities are less than those extrapolated from the observation by Headington that a single FU was observed per each millimeter of occipital scalp which led to the conclusion that average FU density was 100/cm2.(6) It could be surmised that sampling error would account for the difference between calculated density from a single square millimeter compared to counts obtained when a full square centimeter was sampled, including inter-FU spaces. In our pilot study, limited frontal density sampling behind the hairline was observed to average 55 to 60 FU/cm2. Greater population sampling using full square centimeter counts, and taking into consideration ethnic and hair characteristic variations, can provide more accurate information about expected densities. Given that average occipital (“back of head” W.U.) density is only slightly greater than natural frontal densities, the only way to achieve equivalent density would be to harvest comparable areas of donor hair. The limitations of area coverage with such a technique is obvious; harvesting a square centimeter of donor area for every square centimeter which needs to be covered will rapidly outstrip donor supply for all but the most limited patterns of hair loss. Fortunately, we know that such levels of density are not necessary for good to excellent cosmetic results, depending on hair characteristics. Nevertheless, it must be acknowledged for the sake of patients as well as practitioners, that whether or not maximum density can be performed, not withstanding questions of survival or angulation of grafts, in most cases it will not represent natural levels of density seen in patients who do not have hair loss.
The more important and still difficult question to address at this time is what constitutes “cosmetic density” or hair numbers per square centimeter that must be achieved in order to eliminate the visible signs of hair loss. Attempting to achieve equivalent density to what nature originally provided is not possible for advanced patterns, but not necessary either. Based on what we know about the contribution of hair characteristics to visual density, it must be assumed that this number will vary based on characteristics such as hair caliber, curl, and contrast.”
Let me try to put the above discussion simply: In 14 men who had no noticeable MPB: Frontal hairline density averaged 59 FU/cm2, frontal temple hairline density averaged 41 FU/cm2, frontal hair density behind the hairline “averaged 55-60 FU/cm2” and the usually densest fringe hair density at the back of the head ranged from 65 to 85 FU/cm2. Why are these numbers so important? As I mentioned above, while 100 FU/cm2 has for a long time been considered “normal” hair density, that density seems to in fact be wrong. Add to the preceding that most hair transplant surgeons agree that observers of a man’s head do not notice any hair thinning until at least half of the original hair is lost in any area (the individual himself may notice it, but others don’t). Hair restoration surgeons therefore generally aim for no more than half the original hair density in any area because of the limited amount of donor hair we all have. They should be aiming for half of the above-noted numbers rather than half of 100.
Moreover, in my experience, a judicious mixture of 1-hair, 2-hair and 3-hair FU at a density of 20 to 30 FU/cm2, but with varying percentages of 2-hair and 3-hair FU in different areas, nearly always can produce a very satisfying cosmetic hair density if the members of the team that is operating on you are skilled enough to consistently achieve high hair survival rates. (Figures 6 – 10). As noted earlier, this leaves more grafts available for other areas that you may want to treat now or later.
If the patient still has some of his/her original hair left in the recipient area and if the surgeon’s office can consistently produce good hair survival rates, densities of 20-30 FU/cm2 can actually produce quite dense-looking results. This is because you will have a combination of the persisting original hair in addition to the transplanted hair (Figures 11 – 15) (See a proviso on this statement, for men as well as for women in “Transplanting in Female Pattern Hair Loss” below.) Of course as the original hair is lost, the hair density in the treated area will decrease in pace with the rate at which that occurs. But, this may take many years, and for a period of time, while the patient is young, he/she can have the appearance of thick hair. As you age and lose those hairs you can choose to add hair to the previously treated areas if the loss occurs earlier than you would like, but at that point you will be older and maybe you won’t mind having somewhat sparser hair then. In any case you will have a better idea of how big your balding area may become relative to how small your donor area may become. Two other substantial advantages of lower FU densities are, a) that for any given number of grafts, if you use a lower transplanting graft density you can treat a proportionally larger area. For example, if you were planning to use 50 or 60 FU/cm2 and you instead used 25-30 FU/cm2, you could cover twice as large an area at half the originally planned density, b) You will also be more likely to have a lower incidence and severity of “shock loss” if it occurs—temporary loss of some original recipient area hair—because it is caused by recipient area incisions decreasing the blood supply and lower numbers of incisions/cm2 produce less interruption of blood supply/cm2.
And one last thing to consider: The results of lower graft densities produce a more subtle change in your appearance, so that it increases the chance that others will not specifically notice that you have more hair. A large percentage of the patients I have treated that way have told me that people often just say, “You look great!” Or they ask questions like, “Have you lost weight? Are you working out? Have you had a face-lift?” So if you don’t want everybody to notice you’ve had a transplant, you might want to choose closer to 20 FU/cm2 instead of higher graft densities. This works particularly well a) if you still have some hair in the recipient area because you never go from bald to really thick hair b) and is sometimes also preferred by older men who have been bald or almost bald for a considerable time; they just want “some” hair without it being immediately obvious to others that they’ve had a transplant. For example, the patient in Figure 16 had that objective. He said to me, “Do half your usual graft density and twice the area. People will notice it less and if I want to make it thicker once it grows in, I’ll just come back for another transplant in the same areas.” I did 15-20 FU/cm2, he decided he didn’t need any more density to be happy and told me that his friends were just telling him how wonderful he looked without realizing why. Food for thought.
e) Hair Transplanting in Female Pattern Hair Loss (FPHL):
I have written very little about this subject in this correspondence because fortunately I have relatively recently covered it in another publication which can be accessed through my website.(9) The only two things I would emphasize however are a) that the potential donor areas containing permanent hair are typically limited to areas that begin behind the ears (no temple areas), and therefore are smaller in women than in men, as well as lower than in men. As a result, FUE should very rarely be utilized in women if the objective is obtaining the most likely permanent hairs, for the same reasons I have outlined with regard to most young men and b) the more of your original hair in the recipient area, the more slowly and carefully your surgeon must go when making recipient area incisions, in order to make them at the same angle and direction of the original hair. If he/she transects the existing hairs as the incisions are being made he/she may accelerate their permanent loss or even kill them. Quite surprisingly, some doctors don’t believe that, but they are usually the ones who don’t like to operate in hair-bearing areas—whether in men or women because—surprise! they get higher than usual rates of what they mistakenly refer to as only “shock loss” and see very little improvement in these patients. I didn’t realize that some hair restoration surgeons might actually believe that it wouldn’t matter if the hairs were transected or not during the making of the incisions until I had one of them who held that opinion, observe me during surgery. I invited him to my office to watch me work in order to find out if there was anything we did differently that might cause the different results he observed in his patients than I had in mine. After the operation was completed I asked him what he would have done differently during the surgery and he answered that he would have made his incisions more acutely and directed them differently. I replied that I made my incisions in the directions and angles of the existing hair and if I hadn’t done that I would have transected many hairs. His response was “it doesn’t matter”. I was absolutely flabbergasted at that answer and I asked him on what basis he was making that statement. His only reply was to repeat, “it doesn’t matter”. The bottom line is that if you have persisting hair in the area in which you are having your hair loss and you go to a doctor who says that you are not a candidate for transplanting into that area, he/she may be one of those who believe transection doesn’t matter and you should certainly at least get another opinion from another doctor.
Michael Beehner, Jerry Cooley, Paul Cotterill, Eric Eisenberg, Vance Elliott, Ed Epstein, Bessam Farjo, Nilofer Farjo, Marcelo Gandelman, John Gillespie, Robert Haber, James Harris, Victor Hasson, Sungjoo Hwang, Francisco Jimenez, Sheldon Kabaker, Sharon Keene, Russell Knudsen, Bobby Limmer, Antonio Mangubat, Mario Marzola, Jennifer Martinick, Paul McAndrews, Melvin Mayer, Bernie Nusbaum, William Parsley, Damkerng Pathomvanich, David Perez-Meza, Marcelo Pitchon, Carlos Puig, William Rassman, William Reed, Paul Rose, Ron Shapiro, Arthur Tykocinksi, Robin Unger, Walter Unger, James Vogel, Jerry Wong.
1. Unger W., Unger R. Estimating the Number of Lifetime Follicular Units: A Survey and Comments of Experienced Hair Transplant Surgeons. Dermatol Surg May 2013; 39(5), 755-760.
2.Cole J. Mathematics of follicular unit transplantation. Sixth annual meeting of the International Society of Hair Restoration Surgery, Washington, DC, September 16-20, 1998.
3. Unger W., Donor Area Boundaries. In: Hair Transplantation 3rd edition, Unger W, ed., Marcel Dekker Inc., New York, 1995, 183-187.
4. Shapiro R., Implications of a limited total donor supply. In: Hair Transplantation 4th edition, Unger W, Shapiro R, eds., Marcel Dekker Inc., New York, 2004, 439-440.
5. Keene S. Cosmetic Density. In: Hair Transplantation 5th edition, Unger W., Shapiro R., Unger R., Unger M, eds., Informa Healthcare London, New York, 2011, 165-168.
6. Headington JT., Microscopic anatomy of the human scalp. Arch Dermatol 1984;120:449-56.
7. Jimenez F, and Unger W., A Practical Approach to the Donor Area. In: Hair Transplantation 5th edition, Unger W, Shapiro R, Unger R, Unger M., eds., Informa Healthcare London, New York, 2011, 262-267.
8. Bernstein RM, Rassman WR. The logic of follicular unit transplantation. Dermatol Clin 1999;7(2):277-96.
9. Unger W., Candidacy of females for hair transplantation. Hair Transplant Forum International, Volume 21, Number 4, July/August 2011, Cover page and pgs 110-112.