Volume 20

Number 6
June 1989 Allergic reactions in patients patch tested

gy of contact dermatitis. Trans St. Johns Hosp Derrnatol 10. Adams RM, Maibach HI, with members or" North
Soc 1969;55:17-35. American Contact Dermatitis Group. A five-year study
9. Eierman H J, Larsen W, Maibach HI, et al. Prospective of cosmetic reactions. J AM ACAO DERMATOL 1985;
study of cosmetic reactions: 1977-1980. J AM ACAD 13:1062-9.
DERMATOL 1982;6:909-17.

III I I

Essential fatty acids in clinical dermatology

David F. Horrobin, DPhil, BM Kentville, Nova Scotia, Canada

A deficiency of essential fatty acid intake can produce severe cutaneous abnormalities but
is exceedingly rare in clinical practice. Recent research has shown that abnormalities in
essential fatty acid metabolism may play a role in atopic eczema, acne, and psoriasis.
Therapeutic innovations have already resulted from this knowledge, and more are likely to
emerge. (J AM ACAD DERMATOL1989;20:1045-53.)

For almost 60 years it has been known that (GLA), nor DGLA to arachidonic acid, although it
essential fatty acids (EFAs) are necessary for skin can convert GLA to DGLA readily. The epidermis
function. That EFAs may play a role in several must therefore obtain its linolenic acid, its G L A or
common skin disorders has begun to be widely DGLA, and its arachidonic acid separately from
appreciated only in the last decade. the blood. The linoleic metabolites must be made in
An outline of E F A biochemistry is shown in Fig. other tissues, primarily the liver? 3 Because the
1. There are two types of EFAs, the n-6 series epidermis turns over rapidly, there are no substan-
derived from dietary linoleic acid and the n-3 series tial local stores of GLA, D G L A , or arachidonic
derived from oMinolenic acid. Both appear to be acid on which it can draw. The epidermis is
metabolized by the same or a closely related therefore dependent on the continual formation of
enzyme sequence. All EFAs can be converted to a GLA, DGLA, and arachidonic acid by the liver
variety of products by lipoxygenases, and three and on their transport to the skin by the blood. 1-3
EFAs---dihomogamma-linolenic acid (DLGA), The first hepatic step in the metabolism of linoleic
arachidonic acid, and eicosapentaenoic acid acid, 6-desaturation to GLA, has been shown to be
(EPA)--can be converted to prostaglandins and highly vulnerable to a variety of factors that may
thromboxanes by cyclooxygenases. Most work on affect the condition of the skin. Aging; diabetes
EFA metabolites has been done on arachidonic meUitus; high alcohol intake; catecholamines
acid, and an outline of its metabolic products is released during stress; and diets rich in simple
shown in Fig. 2. There is increasing interest in the sugars, trans-fatty acids, or saturated fats are
products derived from D G L A and EPA, howev- known to impair conversion of dietary linoleic acid
er. to DGLA and arachidonic a c i d : ,5 Some of the
The metabolism of EFAs in the skin is different known cutaneous consequences of these factors
than in most other tissues (Fig. 1). The epidermis could relate to their impairment of the supply of
lacks both the 5- and 6-desaturase enzymes ~,2 and EFAs to the skin.
cannot convert linoleic acid to 3,-linolenic acid The n-6 EFAs seem to be considerably more
important in the skin than t h e n-3 EFAs. It is
difficult to demonstrate any specific effects of n-3
From the Efamol Research Institute.
E F A deficiency or treatment, and n-3 EFAs alone
Accepted for publicationJuly 11, 1988.
do not reverse cutaneous features of E F A deficien-
Reprint requests: D. F. Horrobin, DPhil, BM, Efamol Research
Institute, P.O. Box 818, Kentville,Nova Scotia B4N 4H8, Cana- cy?. 6-8 Each of the individual n-6 EFAs may have
da. some specific effects, but these are difficult to
1045
 Journal of the
American Academy of
1046 Horrobin Dermatology

n- 6 FATTY A C I D S n - 3 FATTY A C I D S

LINOLEIC ALPHA-LINOLENIC
B

•L- 6-desaturation
GAMMA-LINOLENIC
absent in skin
-qr

STEARIDONIC
Elongation - present in skin

DIHOMOGAMMALINOLENIC EICOSATETFL~,ENOIC

• L -
ARACHIDONIC
5-desaturation - absent in skin Jr
EICOSAPENTAENOIC
Fig. 1. Outline of metabolism of n-6 and n-3 essent[al fatty acids. Dietary linQleic and
a-linolenic acids can be metabolized along the entire pathway in liver and some other
tissues. In skin, however, 6-desaturation and 5-desaturation steps are absent.

evaluate because of the ongoing metabolism of 7. Transepidermal water loss is greatly increased. The
administered EFAs. Some of the E F A actions are water barrier depends on incorporation of linoleic
due to the EFAs as such, either as the free acids or acid or a related fatty acid into the glucosylceram-
incorporated into phospholipids, cholesterol esters, ides, especially those of the lamellar bodies of the
or triglycerides. Other effects depend on the con- intercellular spaces of the stratum corneum. LT,~ It
version of the free acids to 5-1ipoxygenase or has recently been suggested that the oxygenated
derivatives of the fatty acids are essential for the
12-1ipoxygenase metabolites or to cyclooxygenase
water-blocking function? 9
metabolites.
With the exception of the cat family, in young
EXPERIMENTALLY INDUCED ESSENTIAL animals all the defects can be corrected by the oral
FATTY ACID DEFICIENCY administration of adequate amounts of linoleic
acid. The n-3 EFAs are ineffective and may worsen
Most knowledge of the cutaneous consequences
the situation with regard to the cutaneous capil-
of EFA deficiency comes from animal studies,
laries if given alone),8
which have been well described in numerous
There is little agreement as to which precise
reviews. 3s,~~The changes in all species are similar.
linoleic metabolite is responsible for each function.
The main observations are as follows:
Attempts have been made to answer this question
1. Hair becomes thin and discolored and may be lost. by topical administration of various metabolites.
2. The epidermis becomes scaly and rough. In severe Topical prostaglandin E2~~ or arachidonic acid ~t,22
deficiency a dermatitis may develop. corrects skin scaliness, whereas linoleic acid and
3. Sebaceous glands hypertrophy. EFAs seem to play a GLA have relatively little effect. 2~'22 On the other
particularly important role in sebaceous glands. Top- hand, linoleic acid and G L A efficiently correct the
ically applied linoleic acid and GLA become concen- water permeability problem, whereas D G L A has
trated at these sites. T M EFA deficiency causes no effect and arachidonic acid and prostaglandins
increased sebum viscosity and hyperkeratosis of the
may worsen the disordered permeability. -']'22 The
sebaceous ducts.13
role of arachidonic acid is not clear because some
4. Hyperproliferation with increased epidermal turn-
ever and increased DNA synthesis occurs. TM investigators have claimed that it too improves the
5. Cutaneous capillaries become weakened and readily permeability barrier function) a Linoleic acid may
rupture? have relatively little effect on wound healing and
6. Normal healing of wounds fails to occur, possibly as capillary fragility, whereas arachidonic acid can
a result of defective collagen formation, ~5,~6 correct these rapidly. 22
Volume 20
Number 6
June 1989 Essential fatty acids in clinical dermatology 1047

ARACHIDONIC ACID

12-1ipoxygenase 5-1ipoxygenase Cyclo-oxygenase

12-HETE LEUKOTRIENE
B4 2 series
prostaglandins
Fig. 2. Outline of metabolism of arachidonic acid to proinflammatory metabolites.
12-HETE, 12-Hydroxy-eicosatetraenoic acids.

ESSENTIAL FATTY ACIDS AND SKIN

that will inhibit phospholipase A2 .26 This has the
INFLAMMATION
effect of keeping arachidonie acid locked into the
Both cyclooxygenase and lipoxygenase metabo- membrane phospholipids and of preventing its
lites of arachidon.ic acid play important roles in conversion to the free form. Thus access of arachi-
inflammation, stimulating blood tic,w, pain donic acid to both lipoxygenase and cyclooxygen-
responses, and chemotaxis? 427 Much less attention ase enzymes is denied, and levels of all metabolites
has been paid to the metabolites of the other EFAs, are reduced.
but both EPA and D G L A are now attracting
CLINICAL ASPECTS OF ESSENTIAL FATTY
research activity.
ACIDS AND SKIN
Eskimos, who consume large amounts of EPA
when eating a traditional diet, have a low incidence A dietary deficiency of EFAs as a cause of skin
of psoriasis and other skin diseases. 28 The EPA is problems in adults is exceedingly rare because
believed to compete with arachidonic acid, inhibi, almost all diets contain adequate amounts of
ring conversion of the latter to harmful metabo- linoleic acid. This may not be true during total
litesY Caution should be exercised in interpreting parenteral nutrition with fat-free solutions, espe-
the results of the work with Eskimos, however, ciaUy because high levels of glucose and insulin
because they appear to lack the enzyme 5-desatu- block the mobilization of EFAs from adipose tissue
rase and thus, unlike other human beings, cannot stores. 37"39This problem is now well understood and
make their own arachidonic acid? ~ This inability, in adults is satisfactorily corrected by intravenous
rather than high EPA intake, may be more impor- infusion of linoleic acid. '~ Topical linoleic acid as
tant in reducing inflammation. 3t sunflower or safflower oil has been effective in
DGLA can be converted within the skin to some cases but not in others. 4~42 It is possible that
prostaglandin E~,32 a substance known to raise cows' milk-based infant formulas, which contain
levels of cyclic adenosine monophosphate. Cyclic far lower levels of EFAs than either human milk or
adenosine monophosphate in turn can inhibit phos- vegetable oil-based formulas, may sometimes pro-
pholipases and so exert anti-inflammatory effects? 3 vide inadequate amounts of EFAs for rapidly
Interestingly, human skin prostaglandin El levels growing infants, This could contribute to infant
decline with age? 4,34 Recently it has been found skin problems because nutritionat deprivation of
that D G L A can be converted by skin to a 15- human infants can lead to a cutaneous EFA
hydroxy derivative, which is a powerful inhibitor of deficiency state. 43,44
5-1ipoxygenase (Fig. 2), an enzyme that generates Most of the current interest surrotmding EFAs
leukotriene B4 and other products known to be and ~he skin relates to the concept that intake of
important in skin inflammation?4-27,25 EFAs is usually adequate, but that abnormalities
The anti-inflammatory effects of topical steroids of EFA metabolism may lead to various skin
are due at least in part to induction of a protein diseases.
 Journal of the
American Academy of
1048 Horrobin Dermatology

Atopie dermatitis by transplacental transport from the mother, the

Hansen, a pediatrician and friend of George and consequences of any metabolic abnormality in the
Mildred Burr, discoverers of the essential fatty infant are likely to be diluted. Nevertheless, the
acids, was the first to suggest that atopic eczema higher the IgE level in cord blood, the higher the
might be due to EFA deficiency.45-47 Hansen concentration of linoleic acid (.p < 0.01), suggest-
reported that the blood of patients with atopic ing a positive relationship between high IgE and
eczema had a low iodine number, indicating low impaired linoleic acid metabolism. In an investiga-
levels of unsaturated fats. When he fed large tion in an African population, breast milk from
amounts of linoleic acid, the linoleic acid levels in mothers with atopy contained more linoleic acid
the blood were normal, but those of arachidonic and significantly less GLA, DGLA, and arachi-
acid remained well below normal. At that time it donic acid than breast milk from control subjects
was not known that linoleic acid was metabolized (Wright S, Bolton C, unpublished observation).
to arachidonic acid. Hansen claimed considerable Thus in three different populations, elevated
therapeutic success, and for a few years linoleic amounts of linoleic acid and reduced amounts of
acid was relatively widely used in North America metabolites have been found in association with
in the treatment of atopic eczema. Unfortunately, atopic eczema.
huge doses were required, and this led to a decline If impaired 6-desaturation is important in atopic
in use and eventually to discontinuation of the eczema, then administering GLA should produce
therapy. clinical benefit by bypassing the block. GLA is
Interest in EFAs and atopic eczema was revived present in human milk and in very small amounts in
in the early 1980s when measurements of EFAs in oats but otherwise is not readily available in foods
the blood of adult patients with atopic eczema were that are usually consumed. It is present in variable
performed with modern techniques. Results quantities in the seed oil of the evening primrose,
showed that the levels of linoleic acid and a- Oenothera biennis? 5 Efamol (Epogam) is the oil
linolenic acid in plasma phospholipids tended to be prepared from varieties of Oenothera specially
above normal, whereas concentrations of all their selected and bred to give oil of constant quality and
6-desaturated metabolites were significantly composition. In a double-blind, crossover investiga-
reduced.48,49 This led to the proposal that 6- tion, oral Efamol proved significantly better than
desaturation was moderately impaired in atopic placebo in treating adults with atopic eczemaY ,57
eczema, with a consequent deficit of EFA metabo- This clinical response was associated with rises in
lites.48.49 Because these metabolites are known to be DGLA, arachidonic acid, and prostaglandin Et in
important in skin function, and because GLA and the plasma58,49 Children who received 1 or 2
prostaglandin El are able to act as mild immuno- gm/day showed some improvement but did less well
modulating agents, 33'50-52it was suggested that this than adults. In retrospect, the children should
might be the fundamental defect in atopic ecze- probably have been given a higher dose than the
ma.48,49.53If it were, this would explain the relative- adults. Growing animals have far higher EFA
ly weak therapeutic effect of linoleic acid, because requirements than adults, s8 and the same probably
the conversion to GLA would be restricted. applies to human beings. A second controlled study
Additional biochemical support for this concept in children, using 3 gin/day, produced an excellent
has been obtained recently. A Swedish group has response in the majority of the patients (Bordoni A,
reported exactly the same pattern in young chil- Biagi PL, Cipolla IM, unpublished observation).
dren with atopic dermatitis as previously observed Trials in atopic eczema in adults in Finland and
in British adults: 4 Linoleic acid levels in plasma Germany have produced outcomes similar to those
lecithin were significantly elevated as compared to of the Wright/Burton study.59.6oThese three investi-
those for normal control subjects, whereas levels of gations (i.e., the Wright/Burton study ~7 and the
linoleic metabolites were significantly reduced. studies of MeigeP9 and Schalin-Karrila et al.6o), to-
Because high IgE levels in cord blood may be gether with placebo-controlled trials from five other
indicators of risk of atopic eczema, the Swedish centers, have been combined recently for overall
investigators collected blood from the umbilical analysis. 61 Results showed that patients assessed
cord and measured EFA and IgE concentrations. Efamol (Epogam) as highly significantly better
Because the infant receives much of its EFA supply than placebo. Physicians also assessed Efamol as
Volume 20
Number 6
June 1989 Essential fatty acids in clinical dermatology 1049

significantly better than placebo. There was no pla- Levels of linoleic acid in sebum decrease pro-
cebo response for itch, and the effect of Efamol on gressively as the severity of ache increases.~,67 As
itch was highly significantly better than that of pla- boys go through puberty, linoleic acid in sebum
cebo. 61The combined study also showed that clinical decreases, and the decrease is related to the rise in
responses were confined to those who showed in- the number of a c n e l e s i o n s . 68,69 The amount of
creased levels of the GLA metabolites, DGLA and linoleic acid in each sebaceous cell at the start of
arachidonic acid, in plasma. Failure to show such a differentiation seems to be constant, and therefore
rise, whether because of noncompliance, lack of its concentration will decrease if there are increased
absorption, or other factors, was associated with a rates of synthesis of other lipidsY. 70
failure of clinical effect. Improvement was unex- It is suggested that a local abnormally low
pectedly and particularly associated with a rise in concentration of linoleic acid in sebum may
phospholipid arachidonic acid, suggesting that at produce a local EFA deficiency of the cells of the
least in this form arachidonic acid has an antieeze- follicular epithelium, leading to hyperkeratosis.
ma action and is not pminflammatory. This also will impair the permeability barrier,
There have been three negative reports on GLA allowing greater potential for the penetration of
and atopic eczema. In one, topical administration organisms and proinflammatory fatty acids from
of Efamol was ineffective. 6~ This may be because sebum and promoting infection and inflammation.
skin can convert GLA only to DGLA and not to All known nonantibiotic antiacne agents reduce
arachidonic acid. In an open study on eight patients sebum formation and so are likely to increase the
Skogh 63 found 25% to 35% improvement in dry- concentration of linoleic acid in sebum. This has
ness, itching, and general condition but attributed been shown to occur in practice for cis-retinoic acid
these results to a placebo effect. Bamford et al.,64in and for the cyproterone/estradiol combination.71,72
a placebo-controlled study, failed to observe It is important to appreciate that such a reduced
improvement. The patients in the study of Bamford sebum concentration of EFAs could occur in the
et al., however, were less severely affected by atopic absence of any systemic EFA deficiency. In one
eczema than those in the other studies, and many study EFA levels in plasma phospholipids were
did not have a family history of atopy. Moreover, measured in patients with ache. Levels were nor-
most of the patients in the study of Bamford et al. mal except for those of arachidonic acid, which
failed to show any rise in DGLA and arachidonic were significantly reduced. 73
acid levels, raising questions about compliance. This concept of ache is stimulating and promises
The EFA abnormality that is already detectable to lead to developments in treatment. Whether oral
at birth 54 may predispose to sensitization of the or topical administration of EFAs will be required
immune system to various allergens. The skin and which EFAs will work require further investi-
disease would then be partly due to the EFA defect gation.
and partly to abnormal immunologic reactivity.
Once sensitized, the immunologic abnormality Psoriasis
may not be readily reversed. What is required to The skin of patients with psoriasis contains
settle this question is a prospective controlled study increased total lipid and phospholipid and extraor-
from birth in infants with elevated IgE in cord dinarily large amounts of free arachidonic
blood. Half the babies would be given GLA and acid.27,74,75The arachidonic acid seems to be con-
half would be given placebo. The GLA might vetted preferentially to lipoxygenase products with
prevent the development of sensitization and of reduced retlative or absolute metabolism along the
atopic eczema. cyclooxygenase pathway. The high levels of 5-
lipoxygenase products, especially of leukotriene B4,
Aene
are befieved to play an important role in pathogen-
EFA deficiency produces sebaceous gland esis. Steroids may produce symptomatic improve-
hypertrophy and hyperkeratinization of the ducts, ment by inhibiting phospholipase A2 and so reduce
changes of obvious relevance to acne. 9,~' Only the availability of free arachidonic acid. Although
recently, however, have Downing et al. 6s developed this effect would reduce the quantity of lipoxygen-
the concept that acne may be due to a local deficit ase products, it might weIl worsen any prostaglan-
of linoleic acid in sebum. din deficit. Nonsteroidal anti-inflammatory agents
 Journal of the
American Academy of
1050 Horrobin Dermatology

Skin problems related to age, diabetes mellitus,

that selectively inhibit cyclooxygenase certainly
alcoholism, and steroid use
worsen psoriasis. 74,7~ Whether they do this by
reducing prostaglandin levels or by increasing the Because the supply to the skin of all n-6 EFAs
availability of arachidonic acid for the 5-1ipoxygen- other than linoleic acid is dependent on hepatic
ase is unknown. The 5-lipoxygenase inhibitors, conversion of linoleic acid to GLA and beyond, it is
such as the anti-inflammatory benoxaprofen (now an intriguing possibility that aging skin might be
withdrawn from the market because of toxicity), deficient in GLA. A person's capacity to convert
improve psoriasis. 76 [inoleic acid to GLA decreases with age, and this
Before tar therapy, linoleic acid level in plasma might account for some of the dermal manifesta-
phospholipids of patients with psoriasis is normal, tions of aging, including pruritus, scalingl and
but levels of all n-6 EFAs are reduced. This is thinning of the skin. Levels of prostaglandin El,
similar to the picture in eczema. What is different which can be formed in skin from GLA, 32decrease
from eczema, however, is that the levels of n-3 in human skin with aging. 34 This concept is sup-
EFAs are normal?7 After tar treatment, the only ported by the well-known premature aging of the
plasma EFA to change was DGLA, which rose skin that occurs in alcoholics and in patients with
from a level below normal to a level well above diabetes. Inhibition of 6-desaturation occurs in
normal. Results of another study also showed that both diabetes and alcoholism.'.5 Therefore GLA
DGLA level was high in plasma from patients with and DGLA might be particularly useful in skin
psoriasis, whereas levels for linoleic and oMinolenic care preparations for older people. Of course, I am
acids were below normal. TM The elevated DGLA not suggesting that lack of certain EFAs is the sole
level may well have been due to tar treatment. In or even the main contributor to skin aging. Ultra-
the small free fatty acid fraction in blood from violet radiation in everyone, glycosylation of colla-
patients with psoriasis, linoleic and arachidonic gen in diabetic patients, and acetaldehyde-adduct
acid levels were depressed, whereas EPA and formation in alcoholics are all important mecha-
saturates levels were increased, r~ This pattern of nisms.
free fatty acids is also seen in patients with fatty EFA deficiency leads to greatly increased skin
liver, and it is interesting to note that fatty liver is permeability to and absorption of steroids. 83 This
several times more common in patients with psori- seems to be related to the increased transepidermal
asis than in normal persons. 79 water transport. The EFAs might help to reduce
Eskimos appear to be virtually free from psoria- percutaneous absorption and therefore systemic
sis.z8 Because they consume much EPA, and side effects of steroids.
bec~ause EPA competes with arachidonic acid and
Uses in veterinary medicine
reduces conversion of the latter to potentially
inflammatory me.tabolites, attempts have been Dogs and members of the cat family subject to
made to treat psoriasis with concentrated fish oil.29 EFA deficiency develop syndromes very similar to
Two double-blind, placebo-controlled studies in those in other animals, 8.g7 including poor coat
psoriasis have been published recently.8~ Results quality, hair loss, scaling, dandruff, seborrheic
of one showed a significant improvement; those for dermatitis, eczematQus lesions, and failure of
the other did not. It may be that the absence of wound healing. Also, accumulation of an abnormal
psoriasis in Eskimos is.related more to the inability amount of ear wax is seen. Although for many
to make arachidonie acid by 5-desaturation than to years veterinarians have given animals linoleic and
the high EPA intake, al a-linolenic acids for coat condition, the importance
The rise in DGLA levels with tar therapy sug- of 6-desaturation for the provision of the full range
gests ~hat GLA or DGLA might be of v~alue in of EFAs to the skin has not been appreciated.
psoriasis. Tile observation that topical application Recently, administration of supplements contain-
of prostaglandin E, but not of F-a prostaglandins ing EPA and (3LA to dogsand cats has been found
can cause resolution of psoriatic lesions ~upports to improve coat quality and reduce scaling. 88
this view 24.8r A combination of GLA or DGLA Members of the cat family are unusual in having
with coal tar might produce interesting results. lost the ability to 6-desaturate EFAs.86.87 Cats
Volume 20
Number 6
June 1989 Essential fatty acids in clinical dermatology 1051

therefore cannot survive without direct consump- 8. Kramar J, Levine VE. Influence of rats and fatty acids on
tion of desaturated EFAs, which, as obligate the capillaries. J Nutr 1953;50:149-60.
9. Sinclair HM. Essential fatty acids and the skin. Br Med
carnivores, they get from meat. As with vitamin C Bull 1958;14;258-62.
in human beings, the members of the cat family 10. Sherertz EF. The skin in essential fatty acid deficiency.
may have adapted to managing with lower levels of In: Roe DA, ed. Nutrition and the skin. New York: Alan
R Liss, 1986:117-30.
desaturated EFAs than are required in other 11. Roguet R, LQtte C, Berrebi C, et al. In vivo distribution of
species. Deficiency of desaturated EFAs in cats, linoleic acid in hairless rat skin following topical adminis-
however, m a y be more common than has hitherto tration. Arch Dermatol Res 1986;278:503-6.
12. Costedoat M, Rolez S, Wepierre J. Percutaneous absorp-
been suspected. Cheetahs respond dramatically to tion and distribution of carbon 14--labelled gamma-lino-
G L A supplements with regard to coat condition, lenic acid in hairless rat and man. Int J Cosmetic Sci
fertility, and general h e a k h ? 7 1981;3:83-94.
13. Thiers H, Faycelle J, Divry P. Actions of experimental
Burns nutritional deficiencieson the sebaceous glands of animals
and humans. J Med Lyon 1970;51:1831-44.
Burned skin contains greatly reduced amounts 14. McCullough JL, Schreiber SH, Ziboh VA. Cell prolifer-
of linoleic acid, and the high permeability of such ation kinetics of epidermis in the essential fatty acid
deficient rat. J Invest Dermatol 1978;70:318-20.
skin to water and other agents has been attributed 15. Caffrey BB, Jonsson HT. Role of essential fatty acids in
to E F A deficiency?9~9' It has been suggested that cutaneous wound healing in rats. Prog Lipid Res 1981;
topical E F A s m a y play major roles in the treat- 20:641-7.
16. Hulsey TK, O'Neill JA, Neblett WR. Experimental
ment of burns. wound healing in essential fatty acid deficiency. J Pcdiatr
Surg 1980;15:505.
CONCLUSION 17. Wertz PW, Downing DT. Glycolipids in mammalian
epidermis: structure and function of the water barrier.
Ten years ago the dermatologist could afford to
Science 1982;217:1261-2.
know nothing about EFAs. It now appears that 18. Elias PM. The essential fatty acid deficient rodent:
three of the most common skin disorders, atopic evidence for a direct role for intercellular lipid in barrier
eczema, acne, and psoriasis, m a y be related to function. Models in Dermatol 1985;1:272-85.
19. Nugteren DH, Christ-Hazelhof E, van der Beck A, et al.
abnormalities o f E F A metabolism. It seems proba- Metabolism of linoleie acid and other essential fatty acids
ble, given the importance of the EFAs to the skin, in the epidermis of the rat. Biochem Biophys Acta 1985;
that EFAs will be found to play major or minor 834:429-36.
20. Ziboh VA, Hsia SL. Effects of prostaglandin E2 on rat
roles in several other dermatologic problems. skin: inhibition of sterol ester biosynthesis and clearing of
scaly lesions in essential fatty acid deficienoy.J Lipid Res
REFERENCES 1972;13:458.
1. Chapkin RS, Ziboh VA. Inability of skin enzyme prepa- 21. Hartop PJ, Prottey C. Changes in trans-epiderma[ water
rations to biosynthesize arachidonic acid from linoleic loss and the composition of epidermal lecithin after
acid. Biochem Biophys Res Commun 1984;124:784-92. applications of pure fatty acid triglycerides to the skin of
2. Chapldn RS, Ziboh VA, Marcelo CL, et al. Human essential fatty acid deficient rats. Br J Dermatol 1976;
epidermal preparations can elongate gamma-linolenic 95:255-64.
acid into gamma-homolinolenic acid. J Invest Dermatol 22. Prottey C. Investigation of functions of essential fatty
1986;86:468. acids in the skin. Br J Dermatol 1977;97:29-38.
3. Ziboh VA, Chapkin RS. Biologicsignificanceof polyun- 23. Houtsmuller UMT, van der Beck A. Effects of topical
saturated fatty acids in the skin. Arch Dermatol 1987; application of fatty acids. Prog Lipid Res 1981;20:219-
123:1686a-90. 24.
4. Horrobin DF. The regulation of prostaglandin biosynthe- 24. Kassis V. The prostaglandin system in human skin. Dan
sis by the manipulation of essential fatty acid metabolism. Med Bull 1983;30:320-43.
Rev Pur Appl Pharmacol Sci 1983;4:339-84. 25. Dowd PM, Greaves MW. Cutaneous responses to lipoxy-
5. Brenner RR. Nutritional and hormonal factors influ- genase products of arachidonic acid. Acta Derm Venereol
encing desaturation of essential fatty acids. Prog Lipid [suppl] (Stoekh) 1985;120:18-22.
Res 1982;20:41-8. 26. Camp RDR. Prostaglandins, hydroxy fatty acids, leuko-
6. Burr GO, Brown JB, Kass JP, et al. Comparative curative trienes and inflammation of the skin. Clin Exp DermatoI
values of unsaturated fatty acids in fat deficiency. Proc 1982;7;435-44.
Soc Exp Biol Med 1940;44:242-4. 27. Yardley H J, Summerly R. Lipid composition and metab-
7. Greenberg SM, Calbert CE, Savage EE, et al. The effect olism in normal and diseases epidermis. Pharmacol Ther
of fat level of the diet on general nutrition. VI. The 1981;13:357-83.
inter-relation of linoleate and linolenate in supplying the 28. Kromann N, Green A. Epidemiological studies in the
essential fatty acid requirement in the rat. J Nutr Upernavik District, Greenland. Acta Med Scand 1980;
1950;41:473-86. 208:401-6.
 Journal of the
American Academy of
1052 Horrobin Dermatology

29. Ziboh VA, Cohen KA, Ellis CN, et al. Effects of dietary 48. Manku MS, Horrobin DF, Morse N, et al. Reduced levels
supplementation of fish oil on neutrophil and epidermal of prostaglandin precursors in the blood of atopic patients:
fatty acids. Arch Dermatol 1986;122:1277-82. defective delta-6-desaturase function as a biochemical
30. Gibson RA, Sinclair AJ. Are Eskimos obligate carni- basis for atopy. Prostaglandins Leukotrienes Med 1982;
vores? Lancet 1981;1:1100. 9:615-28.
31. Horrobin DF. Low prevalences of coronary heart disease, 49. Manku MS, Morrobin DF, Morse NL, et al. Essential
psoriasis, asthma and rheumatoid arthritis in Eskimos: are fatty acids in the plasma phospholipids of patients with
they caused by high dietary intake of eicosapentaenoic atopic eczema. Br J Dermatot 1984;110:643-8.
acid, a genetic variation of essential fatty acid metabolism 50. Horrobin DF, Manku MS, Oka M, et al. The nutritional
or a combination of both? Mad Hypotheses 1987;22: regulation of T lymphocyte function. Med Hypotheses
421-8. 1979;5:969-85.
32. McCreedy C, Wong H, Ziboh VA. Generation of prosta- 51. Stackpoole A, Mertin J. The effect of prostaglandin
glandins of the ELseries from dihomogammalinolenic acid precursors on in vivo models of cell-mediated immunity.
by guinea pig epidermal enzyme preparations. J Invest Prog Lipid Res 1981;20:649-54.
Dermatol 1987;88:506. 52. Mertin J, Stackpoole A. Anti-PGE antibodies inhibit in
33. Horrobin DF. The regulation of prostaglandin biosynthe- vivo development of cell-mediated immunity, Nature
sis: negative feedback mechanisms and the selective 1981;294:456-8.
control of formation of 1 and 2 series prostaglandins: 53. Wright S. Atopic dermatitis and essential fatty acids: a
relevance to inflammation and immunity. Mad Hypothe- biochemical basis for atopy. Acta Derm Venereol [suppl]
ses 1980;6:687-709. (Stockh) 1985;114:143-5.
34. Kassis V, Sondergaard J. PGE~ in normal skin: method- 54. Strannegard l-L, Svennerholm L, Strannegard O. Essen-
ological evaluation, topographical distribution and data tial fatty acids in serum lecithin of children with atopic
related to sex and age. Arch Dermatol Res 1983;275: dermatitis and in umbilical cord serum of infants with
9-13. high or low [gE levels. Int Arch Allergy Appl Immunol
35. Miller CC, Ziboh VA, Jones AD. Guinea pig epidermis 1987;82:422-3.
synthesizes 15-hydroxy-8,1I, 13-eieosatfienoic acid 55. Hudson BJF. Evening primrose (Oenothera spp) oil and
(15OH-20:3n-6) from dihomogammalinolenic acid: a seed. J Am Oil Chem Soc 1984;61:540-3.
potent lipoxygenase inhibitor derived from dietary prim- 56. Lovell CR, Burton JL, Horrobin DF. Treatment of
rose oil. J Invest Dermatol 1987;88:507. atopic eczema with evening primrose oil. Lancet 1981;
36. Flower RJ. The mediators of steroid action. Nature 1:278.
1986;320:20. 57. Wright S, Burton JL. Oral evening primrose seed oil
37. Riella MC, Broviac JW, Wells M, et al. Essential fatty improves atopic eczema. Lancet 1982;2:1120-2.
acid deficiency in human adults during total parenteral 58. Aaes-Jorgensen E, Leppik EE, Hayes HW, et al. Essen-
nutrition. Ann Intern Mad 1975;83:786-9. tial fatty acid deficiency. II. In adult rats. J Nutr 1958;
38. Presser SE, Bland H, Ziboh VA. Essential fatty acid 66:245-59.
deficiency after intestinal bypass. J AM ACADDERMATOL 59. Meigel W. Treatment of atopic dermatitis. Z Hautkr
1983;8:127-8. 1986;61:473-8,
39. Richardson J, Sgoutas D. Essential fatty acid deficiency 60. Schalin-Karrila M, Mattila L, Jansen CT, et al. Evening
in four adult patients during total parenteral nutrition. primrose oll in the treatment of atopic eczema: effect on
Am J Clin Nutr 1975;28:258-63. clinical status, plasma phospholipid fatty acids and circu-
40. Friedman Z, Sochot S, Maisels M, et al. Correction of lating blood prostaglandins. Br J Dermatol 1987;117:
essential fatty acid deficiency in newborn infants by 11-9.
cutaneous application of sunflower-seed oil. Pediatrics 61. Morse PF, Horrobin DF, Manku MS, et al. Meta-
1976;58:650-4. analysis of placebo-controlled studies of the efficacy of
41. Hunt CE, Engle RR, Modler S, et al. Essential fatty acid gamma linoleic acid in the treatment of atopic eczema.
deficiency in neonates: inability to reverse deficiency by Relationship between plasma essential fatty acid changes
topical applications of EFA-rich oil. J Pediatr 1978; and clinical response. Br 3 Dermatol [In press].
92:604-7. 62. Macdonald KJS, Green C, Raffle E J, et al. Topical
42. Miller DG, Williams SK, Palombo JD, et al. Cutaneous evening primrose seed oil and atopic eczema. Scott Med J
application of safflower oil in preventing essential fatty 1985;30:267.
acid deficiency in patients on home parenteral nutrition. 63. Skogh M. Atopic eczema unresponsive to evening prim-
Am J Clin Nutr 1987;46:419-23. rose oil (linoleic and gamma-linolenic acids). J AM ACAD
43. Hansen AE. Role of unsaturated dietary fat in infant DERMATOL1986;15:114-5.
nutrition. Am J Public Health 1957;47:1367-70. 64. Bamford JTM, Gibson RW, Renier CM. Atopic eczema
44. Hansen AE, Haggard ME, Boelsche AN, et al. Essential unresponsive to evening primrose oil (linoleic and gamma-
fatty acids in human nutrition. III. Clinical manifestations linolenic acids). J AM ACADDERMATOL1985;3:959-65.
of linoleic acid deficiency. J Nutr 1958;66:565-76. 65. Downing DT, Stewart ME, Wertz PW, et al. Essential
45. Hansen AE. Serum lipid changes and therapeutic effects fatty acids and acne. J AM ACAD DERMATOL1986;14:
of various oils in infantile eczema. Proc Soc Exp Biol Meal 221-5.
1933;31:160-1. 66. Morello AM, Downing DT, Strauss JS. Octadecadienoic
46. Hansen AE. Serum lipids in eczema and other patholog- acids in the skin surface lipids of ache patients and normal
ieal conditions. Am J Dis Child 1937;53:933-46. subject. J Invest Dermatol 1976;66:319-23.
47. Brown WR, Hanse AE. Arachidonic acid and linoleic 67. Downing DT, Stewart ME, Wertz PW, et al. Skin lipids.
acid of serum in normal and eczematous human subjects. Comp Biochem Physiol 1983;76B:673-8.
Proc Sor Exp Biol Mad 1937:30:113-6. 68. Bazzano GS, Cummings B, Seeler AK, et al. Differences
Volume 20
Number 6
June 1989 Essential fatty acids in clinical dermatology 1053

in the lipid constituents of sebum from prepubertal and 80. Bittiner SB, Tucker WFG~ Cartwright I, et al. A double-
pubertal subjects. Br J Dermatol 1980;103:131-7. blind, randomized, placebo-controlled trial of fish oil in
69. Bazzano GS, Reisner RM, Cummings B, et al. Prelimi- psoriasis. Lancet 1988;1:378-80.
nary observations of a prospective longitudinal study of 81. Bjorneboe A, Smith AK, Bjorneboe G-EAA, et al. Effect
the development of ache in children. Semin Dermatol of dietary supplementation with n-3 fatty acids on clinical
1982;1:315-8. manifestations of psoriasis. Br J Dermatol 1988;118:77-
70. Stewart ME, Greenwood R, Cunliffe W J, et al. Effect of 83.
cyproterone acetate-ethinyl estradiol treatment on the 82. Jacobs KF, Jacobs MM. Pr.ostag[andin treatment of
proportions of linoleic and sebaleic acids in various skin psoriatic skin. Rocky Mt Med J 1974;7I:507-10.
surface lipid classes. Arch Dermatol Res 1986;278: 83. Lambrey B, Daniel MH, Lahmy JP, et al. Study of the
48 I-5. cutaneous penetration of hydrocortisone in the hairless rat
71. Stewart ME, Grahek MO, Cambier LS, et al. Dilutional deficient in essential fatty acids. J Pharm Clin 1985;
effect of increased sebaceous gland activity on the propor- 4:203-10.
tion of linoleie acid in sebaceous (amino) wax esters and in 84. Hansen AE, Wiese HF. Studies with dogs maintained on
epidermal acyl-ceramides. J Invest Dermatol 1986; diets low in fat. Proe Soc Exp BiN Med 1943;52:205-8.
87:733-6. 85. Wiese HF, Yamanaka W, Coon E, et al. Skin lipids of
72. Stewart ME, Sears JK, Strauss JS, et al. Increased puppies as affected by kind and amount of dietary fat. J
proportions of linoleate in human sebaceous lipids during Nutr 1966;89:113-22.
treatment with 13-cis-retinoic acid. Clin Res 1986; 86. Frankel TL, Rivers JPW. The nutritional and metabolic
34:783A. impact of gammalinolenic acid on cats deprived of animal
73. Grattan C, Burton JL, Manku MS, et al. Essential fatty lipid. Br J Nutr 1978;39:227-31.
acids in plasma phospholipids in patients with ichthyosis, 87. Davidson BC, Cantrill RC, Varaday D. The reversal of
acne and psoriasis. Br J Dermatol [In press]. essential fatty acid deficiency symptoms in the cheetah. S
74. Kragballe K, Voorhees JJ. Arachidonic acid and leuko- Aft J Zool 1986;2l:16l-4.
trienes in clinical dermatology. Curt Probl Dermatol 88. Lloyd D. Dietary supplementation in the treatment of
1985;13:1-10. veterinary skin disease. Presented at the 5th Congress of
75. Ellis CN, Gorsulowsky DC, Voorhess JJ. Experimental the European Society of Veterinary Dermatology, Lon-
therapies for psoriasis. Semin Dermatol 1985;4:414-19. don, England, April 7-10, 1988; abstract book, pp 77-
76. Kragballe K, Herlin T. Benoxaprofen improves psoriasis: 81.
a double blind study. Arch Dermatol 1983;119:548-52. 89. Jelenko C, McKinley JC. Studies in burns. XV. Use of a
77. Strong AMM, Horrobin DF, Manky MS, et al. Prosta- topical lipid in treating human burns. Am Surg 1976;
glandin precursors in plasma phospholipids of patients 42:838-48,
with psoriasis: effects of treatment with coal tar. Prosta- 90. Jelenko C, Wheeler ML, Scott TH. Studies in burns. X.
glandins Leukotrienes Med 1984;14:185-90. Ethyl linoleate: the water-holding lipid of skin. A. The
78. Vahlquist C, Berne B, Boberg M, et al. The fatty acid evidence. J Trauma 1972;12:968-73.
spectrum in plasma and adipose tissue in patients with 91. Shakespeare PG, Strange RE. Linoleic acid in hyper-
psoriasis. Arch Dermatol Res 1985;278:114-9. trophic scars. Burns Incl Therm lnj 1982;9:7-12.
79. Zlatkov NB, Ticholov JT, Dourmishev AL. Free fatty
acids in the blood stream of psoriatics. Acta Derm
Venereol (Stock_h) 1984;64:22-5.

BOUND VOLUMES AVAILABLETO SUBSCRIBERS

Bound volumesof the JOURNALOFTHEAMERICANACADEMYOFDERMATOLOGYare available to subscribers (only) for the 1989 issues from
the Publisher at a cost of $52.00 ($65.00 international) for volume20 (January-June) and volume21 (July-December).Shippingcharges are
included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after
publication of the last issue in the volume.The binding is durable buckram with the journal name, volumenumber, and year stamped in gold
on the spine. Payment must accompany all orders. Contact The C.V. Mosby Company,Circulation Department, 11830Westline Industrial
Drive, St. Louis, MO 63146-3318. USA: phone (800) 325-4177, ext. 351.
Subscriptions must be in force to qualify. Bound vol,tmes are not available in place o f a regular journal subscription.