You are on page 1of 10



Atopic dermatitis

Donald Y M Leung, Thomas Bieber

Atopic dermatitis is a highly pruritic chronic inflammatory skin disorder affecting 10–20% of children worldwide.
Symptoms can persist or begin in adulthood. It is also the most common cause of occupational skin disease in adults.
This disease results from an interaction between susceptibility genes, the host’s environment, pharmacological
abnormalities, skin barrier defects, and immunological factors. New management approaches have evolved from
advances in our understanding of the pathobiology of this common skin disorder.

Atopic dermatitis is a highly pruritic chronic inflammatory Th1 responses antagonise development of Th2 cells, a
skin disease that commonly presents during early infancy decreased number of infections or the absence of Th1
and childhood but can persist or start in adulthood.1 polarising signals (such as endotoxin) during early
Interest in this disease has been sparked by reports of its childhood could predispose children to enhanced Th2
increasing prevalence and the significant adverse effects it allergic responses.
can have on quality of life. At least two types of atopic
dermatitis have been identified: an extrinsic type Clinical diagnosis
associated with IgE-mediated sensitisation, which affects Atopic dermatitis offers a wide clinical spectrum ranging
70–80% of patients; and an intrinsic type without IgE- from minor forms such as pityriasis alba (dry depigmented
mediated sensitisation, which affects 20–30% of patients.2 patches) or hand eczema to major forms with
Here, we will discuss some of the advances in our erythrodermic rash. The most common forms include the
understanding of this disease and new treatment clinical features listed in panel 1.9 Of the major features,
approaches that have been introduced. pruritus and chronic or relapsing eczematous lesions with
typical shape and distribution are essential for diagnosis.
Epidemiology Although pruritus can occur throughout the day, it is
Atopic dermatitis is a major public-health problem usually worse in the early evening and night. Pruritis
worldwide with a lifetime prevalence in children of results in scratching, lichenification, and prurigo papules.
10–20%, and a prevalence of 1–3% in adults.3 Prevalence Patients with atopic dermatitis have a reduced threshold
of this disease has increased by two to three-fold during the for pruritus. Allergens, reduced humidity, excessive
past three decades in industrialised countries, but remains sweating, and low concentrations of irritants can
much lower in agricultural regions such as China, eastern exacerbate pruritus and scratching.
Europe, and rural Africa. Moreover, higher prevalences Acute and subacute skin lesions (figure 1) are often
have been recorded in urban regions than in rural regions seen in children and are characterised by intensely pruritic
of developed countries, and the disease is more common in erythematous papules associated with excoriation and
higher social class groups.4 serous exudate. Chronic atopic dermatitis is characterised
Wide variations in prevalence have been identified by lichenification, papules, and excoriations. At all stages
within countries inhabited by similar ethnic groups, of this disease, patients usually have dry lacklustre skin.
suggesting that environmental factors determine The distribution and skin reaction pattern varies
expression of atopic dermatitis.5 Results of comparative according to the patient’s age and disease activity. During
studies in former east and west Germany have confirmed infancy, atopic dermatitis is generally more acute and
that lifestyle and environment play a major part in mainly affects the face, scalp, and extensor surfaces of the
expression of atopic diseases, including atopic dermatitis.6 extremities. In older children and in those who have long-
Some of the risk factors associated with the rise in atopic standing skin disease, the patient develops lichenification
disease include small family size, increased income and and localisation of the rash to the flexural folds of the
education, migration from rural to urban environments, extremities. Chronic hand eczema can be the primary
and increased use of antibiotics—ie, the so-called western manifestation of many adults with atopic dermatitis.
lifestyle.7 These observations are lent support by studies8 in Panel 2 lists other disorders that share symptoms and
which allergic responses were shown to be driven by signs with atopic dermatitis. These should be considered
T-helper type (Th) 2 immune responses, whereas and ruled out before a diagnosis of atopic dermatitis is
infections are induced by Th1 immune responses. Since made.

Lancet 2003; 361: 151–60 Search strategy and selection criteria

We did a computer-aided search of PubMed from 1990
Division of Pediatric Allergy and Immunology, National Jewish
through May, 2002, for aspects of atopic dermatitis pertinent
Medical and Research Center, Denver, CO, USA, and Department
of Pediatrics, University of Colorado Health Sciences Center,
to this review, to supplement our existing awareness of the
Denver, CO (Prof D Y M Leung MD); Department of Dermatology, primary published work. We searched using the keywords
University of Bonn, Bonn, Germany (Prof T Bieber MD) atopic dermatitis and eczema. Because of limitations on the
number of citations, we made selections from the 2043
Correspondence to: Prof Donald Y M Leung, National Jewish Medical
reports published on atopic dermatitis in the past decade to
and Research Center, 1400 Jackson Street, Room K926, Denver,
support our interpretations with criteria for assessing
CO 80206, USA
experimental studies and evidence-based medicine.

THE LANCET • Vol 361 • January 11, 2003 • 151

For personal use. Only reproduce with permission from The Lancet Publishing Group.

Panel 1: Clinical features of atopic dermatitis* Panel 2: Differential diagnosis of atopic dermatitis
Essential features Chronic dermatoses
Pruritus Seborrhoeic dermatitis
Facial and extensor eczema in infants and children Contact dermatitis (allergic or irritant)
Flexural eczema in adults
Nummular eczema
Chronic or relapsing dermatitis
Frequently associated features Ichthyoses
Personal or family history of atopic disease
Congenital disorders
Cutaneous infections Netherton's syndrome
Non-specific dermatitis of the hands or feet Familial keratosis pilaris
Raised serum IgE concentrations Infections and Infestations
Positive immediate-type allergy skin tests Scabies
Early age of onset HIV-associated dermatitis
Other features
Ichthyosis, palmar hyperlinearity, keratosis pilaris Malignant diseases
Pityriasis alba Cutaneous T-cell lymphoma (Mycosis, fungoides/Sézary
Nipple eczema syndrome)
White dermatographism and delayed blanch response Letterer-Siwe disease
Anterior subcapsular cataracts, keratoconus
Immunological disorders
Dennie-Morgan infraorbital folds, orbital darkening
Dermatitis herpetiformis
Facial erythema or pallor
Pemphigus foliaceus
*Other skin disorders that can mimic atopic dermatitis should be Graft versus host disease
excluded. Dermatomyositis
Pathophysiology Wiskott-Aldrich syndrome
Interactions between susceptibility genes, the host’s Severe combined immunodeficiency
environment, pharmacological abnormalities, and Hyper-IgE syndrome
immunological factors contribute to the pathogenesis of DiGeorge syndrome
atopic dermatitis.10 Most of the progress made in Metabolic disorders
understanding the immunology of this disease is related to Zinc deficiency
the IgE-mediated or extrinsic form of the disease. Clearly, Pyridoxine (vitamin B6) and niacin
atopic dermatitis has an immunological basis—as Multiple carboxylase deficiency
confirmed by the observation that primary T-cell Phenylketonuria
immunodeficiency disorders frequently have raised
concentrations of serum IgE and eczematoid skin lesions
that are cleared after successful bone marrow The systemic immune response
transplantation.11 In animals, atopic dermatitis does not Most patients with atopic dermatitis have peripheral
occur in the absence of T cells.12 blood eosinophilia and increased serum IgE
concentrations. Nearly 80% of children with atopic
dermatitis develop allergic rhinitis or asthma, suggesting
that allergen sensitisation through the skin predisposes
people to respiratory diseases.13 Peripheral blood
mononuclear cells from patients with atopic dermatitis
have a decreased capacity to produce interferon ␥, which
is inversely correlated with serum IgE concentrations.
This association could be due to a deficiency of
interleukin 18, an inducer of interferon ␥ production.14
An increased frequency of allergen-specific T cells
producing interleukin 4, 5, and 13, but little interferon ␥,
in the peripheral blood of patients with atopic dermatitis
has also been recorded.15 These immunological changes
are important because interleukins 4 and 13 are the only
cytokines that induce germline transcription at the
⑀ exon, thereby promoting isotype switching to IgE.
These cytokines also induce expression of vascular
adhesion molecules such as VCAM-1, which are
involved in eosinophil infiltration and downregulate
Th1-type cytokine activity. By contrast, interferon ␥
inhibits synthesis of IgE, proliferation of Th2 cells, and
expression of the interleukin 4 receptor on T cells.
Interleukin 5 plays a key part in development, activation,
and survival of eosinophils. The cytokine micromilieu in
Figure 1: Clinical (upper) and histological (lower) findings in which T-cell development takes place, pharmacological
acute and chronic atopic dermatitis factors, the costimulatory signals used during T-cell
Acute (left) and chronic (right) lesions in atopic dermatitis. activation, and the antigen-presenting cell determine the
Magnification ⫻100. outcome of the T-cell response.16–18

152 THE LANCET • Vol 361 • January 11, 2003 •

For personal use. Only reproduce with permission from The Lancet Publishing Group.

Skin pathology and immunopathology naturally occurring lesional skin has significantly more
Clinically unaffected skin of patients with atopic cells expressing mRNA of interleukins 4, 5, and 13.
dermatitis is not normal, but reveals a sparse perivascular However, acute atopic dermatitis does not contain
T-cell infiltrate.19,20 Acute eczematous skin lesions are significant numbers of cells expressing mRNA of
characterised by marked epidermal intercellular oedema interferon ␥ or interleukin 12. By contrast, chronic atopic
(spongiosis) (figure 1). Antigen-presenting cells (eg, dermatitis skin lesions have significantly fewer cells
Langerhans cells, inflammatory dendritic epidermal cells, expressing mRNA of interleukins 4 and 13, but increased
and macrophages) in lesional and, to a lesser extent in numbers of cells expressing mRNA of interleukin 5, GM-
skin without lesions, have IgE molecules (figure 2).21 In CSF, interleukin 12, and interferon ␥ than do those of
dermis with acute lesions, there is a striking infiltration of acute atopic dermatitis.
CD4 activated T cells. Chronic lichenified lesions are In another approach using the atopy-patch test
characterised by an acanthotic epidermis with elongation technique as a model, lesions induced by house dust
of the rete ridges, parakeratosis, and only minimum mites were shown to have two phases: an initial phase
spongiosis (figure 1). Chronic lichenified lesions have an with mostly Th2 cells producing interleukin 4 and a
increased number of IgE-bearing Langerhans cells and subsequent phase after 24–48 h with mainly Th1 cells
inflammatory dendritic epidermal cells in the epidermis, producing interferon ␥.24 This switch is probably started
and macrophages dominate the dermal mononuclear by local production of interleukin 12 from infiltrating
cell infiltrate. These lesions also contain eosinophils, eosinophils, or inflammatory dendritic epidermal cells,
which are thought to contribute to inflammation and or both.
tissue injury through production of reactive oxygen Activated T cells also induce keratinocyte apoptosis,
intermediates and proinflammatory cytokines and release contributing to the spongiotic process seen in acute atopic
of toxic granule proteins. dermatitis.25 This process is mediated by interferon ␥
Cytokine expression—Two approaches have provided derived from T cells, which upregulates Fas on
important information about the cytokine pattern keratinocytes. The lethal hit is delivered to keratinocytes
expressed in acute versus chronic atopic dermatitis by membranous and soluble Fas-ligand produced by
(figure 3). First, analyses of biopsy samples from epidermotropic T cells.
unaffected skin of patients with atopic dermatitis, as Antigen-presenting cells—Atopic dermatitis skin
compared with healthy non-atopic skin, have an increased contains an increased number of IgE-bearing Langerhans
number of Th2 cells expressing mRNA of interleukins 4 cells and inflammatory dendritic epidermal cells
and 13.22,23 When compared with healthy skin or expressing the high-affinity receptor for IgE.26 These
unaffected skin of patients with atopic dermatitis, antigen-presenting cells seem to have an important role in
allergen presentation to Th2 and Th1 cells, respectively.27
In this regard, high-affinity receptors for IgE contribute
to the capture and internalisation of allergens before their
processing and antigen presentation to T cells in atopic
skin. Langerhans cells positive for these receptors can also
migrate to the lymph nodes and stimulate naive T cells,
thereby contributing to expansion of the pool of Th2 cells.
The clinical importance of these cells is supported by the
observation that Langerhans cells with high-affinity
receptors for IgE need to be present to provoke
eczematous skin lesions by application of aeroallergens on
the skin of atopic patients.28
Inflammatory cell infiltration—Interleukin 16, a
chemoattractant for CD4 T cells, is more highly expressed
in acute than chronic atopic dermatitis skin lesions.29,30
The C-C chemokines, T cells expressed and secreted
(RANTES), monocyte chemotactic protein 4 (MCP4),
and eotaxin are also increased in atopic dermatitis skin
lesions and probably contribute to the chemotaxis
of eosinophils expressing CCR3 and Th2 lymphocytes
into the skin.31,32 Cutaneous T-cell-attracting chemokines
could have a role in the preferential attraction of
cutaneous lymphoid antigen T cells to the skin.33 Selective
recruitment of Th2 cells expressing CCR4 into skin with
atopic dermatitis might also be mediated by the
chemokines MDC and TARC, which are increased in
atopic dermatitis.34–36 Persistent skin inflammation in
chronic lesions could be because of expression of
interleukin 5 and GM-CSF in the skin, which lengthen
survival of eosinophils, monocyte-macrophages, and
Langerhans cells.37
Mechanical trauma also induces release of tumour
necrosis factor ␣ and many other proinflammatory
cytokines from epidermal keratinocytes.38 Epidermal
keratinocytes from patients with atopic dermatitis
produce significantly higher concentrations of RANTES
Figure 2: Dendritic cells in atopic dermatitis skin lesion after stimulation with tumour necrosis factor ␣ and
Expression of CD1a (upper) and IgE (lower). Magnification ⫻100. interferon ␥.39 In atopic dermatitis, but not non-atopic

THE LANCET • Vol 361 • January 11, 2003 • 153

For personal use. Only reproduce with permission from The Lancet Publishing Group.

Uninvolved Acute Chronic

Allergens Scratching toxins


IL 13
IL 4
IL 5
ThO IL 12 Th1 IFN ␥

CLA+Th2 cells IgE Eosinophils

Figure 3: Immunological pathways in atopic dermatitis

IL=interleukin. Modified from reference 10 with permission from Mosby/Elsevier Science, St Louis.

dermatitis, keratinocytes are also an important source of chromosome 3q21, a region that encodes the costimu-
thymic stromal lymphopoietin, which activates dendritic latory molecules CD80 and CD86. By contrast, Cookson
cells to prime naive Th cells to produce interleukins 4 and and coworkers46 reported that both atopic dermatitis and
13 and tumour necrosis factor ␣.40 These observations psoriasis were linked to loci on chromosome 1q21 and
might explain the link between scratching and the 17q25, suggesting that common candidate genes are
triggering of Th2-mediated skin inflammation in atopic involved in control of skin inflammation. Since most of
dermatitis. these studies included patients with raised concentrations
of IgE—ie, patients with the IgE-mediated (extrinsic)
Genetics type—the genetic background of the non-allergic
Atopic dermatitis is a genetically complex, familially (intrinsic) type remains elusive. However, since these
transmitted disease with a strong maternal influence. patients usually have the same family history of atopic
Parental atopic dermatitis confers a higher risk to diseases as the classical IgE-mediated form, at least one
offspring than does parental asthma or allergic rhinitis, set of common genes unrelated to IgE-biology is expected.
suggesting the existence of genes specific to atopic
dermatitis. Several chromosomal regions contain Immunological triggers
pathophysiologically relevant candidate genes, especially Foods
on chromosome 5q31–33 since it contains a clustered Food allergens induce skin rashes in nearly 40% of
family of Th2 cytokine genes—ie, interleukins 3, 4, 5, and children with moderate to severe atopic dermatitis.47 Food
13, and GM-CSF.41 Results of a case-control compar- allergies in patients with atopic dermatitis might induce
ison42 have suggested a genotypic association between the dermatitis and contribute to severity of skin disease in
T allele of the –590C/T polymorphism of the interleukin 4 some patients, whereas in others urticarial reactions, or
gene promoter region with atopic dermatitis. That this non-cutaneous symptoms are elicited. Infants and young
allele is associated with increased activity of the children with food allergies generally have positive
interleukin 4 gene promoter suggests that it might affect immediate skin tests or serum IgE directed to various
atopic dermatitis predisposition. Similarly, variants of foods, especially egg, milk, wheat, soy, and peanut.48
the interleukin 13 coding region, a gain-of-function Importantly, T cells specifc to food allergens have been
polymorphism in the ␣ subunit of the interleukin 4 cloned from the skin lesions of patients with atopic
receptor (16q12), and a functional mutation in the dermatitis, providing direct evidence that foods can
promoter region of RANTES (17q11) have a role in contribute to skin inflammation.49 In mice with atopic
expression of atopic dermatitis. Atopic dermatitis has dermatitis, oral sensitisation with foods results in
been controversially linked to markers at chromosome eczematous skin lesions on repeat oral food challenges.50
11q13, including the gene encoding for the ␤ chain of the
high affinity receptor for IgE.43 Atopic dermatitis has also Aeroallergens
been associated with a low-producer transforming growth Pruritus and skin lesions can develop after intranasal or
factor ␤-cytokine genotype,44 which could contribute to bronchial inhalation challenge with aeroallergens in
increased skin inflammation. sensitised (specific IgE) patients with atopic dermatitis.51
Genome-wide linkage studies have also been done to Epicutaneous application of aeroallergens (eg, house dust
identify susceptibility loci for atopic dermatitis. Lee and mites, weeds, animal danders, and moulds) by atopy
colleagues45 suggested that atopic dermatitis is linked to patch test on unaffected atopic skin elicits eczematoid

154 THE LANCET • Vol 361 • January 11, 2003 •

For personal use. Only reproduce with permission from The Lancet Publishing Group.

reactions in 30–50% of patients with atopic dermatitis.52 Th2 inflammatory responses, bacterial binding was
By contrast, atopy patch tests are usually negative in significantly greater at skin sites with Th2-mediated
patients with respiratory allergy and in healthy volunteers. inflammation than at those with Th1-mediated
A combination of effective measures for reduction of inflammation because of expression of fibronectin
house dust mites has been associated with significant induced by interleukin 4. Atopic dermatitis skin is also
improvement in atopic dermatitis.53,54 The degree of IgE deficient in antimicrobial peptides needed for host
sensitisation to aeroallergens is directly associated with the defense against bacteria, fungi, and viruses.72 Thus, once
severity of atopic dermatitis.55 The isolation from skin S aureus binds to atopic dermatitis skin, inadequate host
lesions with atopic dermatitis and allergen patch test sites defence allows bacteria to colonise and predispose
of T cells that selectively respond to Dermatophagoides patients to infection.
pteronyssinus and other aeroallergens provides further
evidence that the immune response in atopic dermatitis Management
skin can be elicited by aeroallergens.52 Successful management of atopic dermatitis requires a
multipronged approach involving skin care, identification
Autoallergens and elimination of flare factors, and anti-inflammatory
In the 1920s, several investigators56 reported that human treatment.73 Randomised controlled trials are especially
skin dander could trigger immediate hypersensitivity important in assessing the effects of treating atopic
reactions in the skin of patients with severe atopic dermatitis because of the substantial placebo effect in this
dermatitis, suggesting that these patients made IgE disease.
against autoantigens in the skin. Most sera from patients
with severe atopic dermatitis contain IgE antibodies Skin care
directed against human proteins.57 Although the In atopic dermatitis, the disturbed function of the skin
autoallergens characterised to date are mainly intracellular barrier is probably the result of reduced ceramide
proteins,58 they have been detected in IgE immune concentrations and results in dry skin (xerosis) and
complexes of atopic dermatitis sera, suggesting that enhanced transepidermal water loss.74 Irritants such as
release of these autoallergens from damaged tissues could soaps or detergents, contact with chemicals, smoke,
trigger responses mediated by IgE or T cells. This notion alcohol and astringents found in toiletries, and abrasive
is supported by the observation that IgE autoallergen clothing can worsen the xerosis. Soaps with minimum
titres decrease with successful treatment of atopic defatting activity and a neutral pH are preferred. Xerosis
dermatitis with ciclosporin,59 suggesting that although IgE contributes to development of epithelial microfissures and
immune responses are initiated by environmental cracks, which allows entry of skin pathogens, irritants, and
allergens, allergic inflammation can be maintained by allergens. Wet dressings can be used on severely affected
human endogenous antigens, especially in severe atopic or chronic lesions refractory to skin care. Dressings can be
dermatitis. an effective barrier against persistent scratching, allowing
more rapid healing of excoriated lesions.
Staphylococcus aureus Every attempt should be made to allow patients to be as
Increased numbers of S aureus are found in over 90% of normally active as possible. Some sports such as
atopic dermatitis skin lesions. The importance of this swimming might be better tolerated than other sports
organism is lent support by the observation that, in involving intense perspiration, and physical contact, but
patients with atopic dermatitis with secondary infection, chlorine should be rinsed off immediately after swimming
treatment with a combination of anti-staphylococcal and the skin lubricated.
antibiotics and topical glucocorticoids produces superior
clinical effects than does treatment with topical Identification and elimination of triggering factors
glucocorticoids alone.60 One strategy by which S aureus Potential allergens can be identified by taking a careful
exacerbates or maintains skin inflammation in atopic history and doing selective allergy tests. Negative skin
dermatitis is by secreting superantigens, which stimulate prick tests or serum tests for allergen-specific IgE have
marked activation of T cells and macrophages.61,62 An a high predictive value for ruling out suspected allergens.
analysis of the peripheral blood skin-homing CLA+ Positive skin or in-vitro allergy tests, especially to
T cells from these patients and T cells in their skin lesions foods, often do not correlate with clinical symptoms and
shows that they have undergone a T-cell receptor V␤ should be confirmed with controlled food challenges,
expansion consistent with superantigenic stimulation.63,64 elimination diets, or atopy patch tests. Avoidance of
Most patients with atopic dermatitis make specific IgE foods implicated in controlled challenges results in
antibodies directed against staphylococcal superantigens; clinical improvement.47,75 As a rule, extensive elimination
these super antigens to IgE correlate with the severity of diets, which in some cases can be nutritionally deficient,
disease.62 Superantigens augment synthesis of allergen- are useless. Most children who are allergic to food
specific IgE and induce glucocorticoid resistance, outgrow their food hypersensitivity in the first few years of
suggesting that superantigens could increase severity of life, making it less relevant as trigger factor when older.
atopic dermatitis through several mechanisms.65,66 Extended avoidance of house dust mites in sensitised
Fulfilling Koch’s postulates, application of atopic patients with atopic dermatitis results in improvement of
dermatitis superantigens to the skin induces dermatitis.67 their skin disease. Avoidance measures include use of
Increased binding of S aureus to skin is probably related house dust mite-proof encasings on pillows, mattresses,
to underlying atopic dermatitis inflammation. This and boxsprings; washing bedding in hot water weekly;
observation is lent support by the finding that treatment removal of bedroom carpeting; and decreasing indoor
with topical glucocorticoids or tacrolimus reduces humidity levels.53,54
S aureus counts on atopic skin.68,69 Scratching probably Unlike allergic rhinitis and extrinsic asthma,
enhances S aureus binding by disturbing the skin barrier immunotherapy with aeroallergens has not been proven to
and exposing extracellular matrix adhesins for S aureus be effective in treatment of atopic dermatitis. Well
such as fibronectin and collagens. In studies70,71 of controlled studies are still needed to determine the role of
S aureus binding to skin lesions undergoing Th1 versus immunotherapy in this disease, and this approach should

THE LANCET • Vol 361 • January 11, 2003 • 155

For personal use. Only reproduce with permission from The Lancet Publishing Group.

be reserved for individuals who have a clear cut Pimecrolimus—Ascomycin compounds such as
demonstrable history of aeroallergen induced atopic pimecrolimus, which has the same mechanism of action as
dermatitis—eg, seasonal exacerbations to pollen. tacrolimus, have been developed in topical and oral forms.
Like tacrolimus, they inhibit production of Th1 and
Topical anti-inflammatory agents Th2 cytokines, and inhibit mediator release from mast
Glucocorticoids—Topical glucocorticoids are frequently cells and basophils.83 1% pimecrolimus is effective and
used to control acute exacerbation of atopic dermatitis. safe in adults and children with atopic dermatitis.84,85
However, once control of atopic dermatitis is achieved Pimecrolimus cream ointment 1% has been approved for
with a daily regimen of topical glucocorticoids, long-term short-term and intermittent long-term use in mild-
control can be maintained with twice weekly applications moderate atopic dermatitis for patients 2 years and older.
of topical fluticasone to regions that have healed but are When used as maintenance therapy, topical pimecrolimus
prone to developing eczema.76 Side-effects from topical reduces the number of flares caused by atopic dermatitis
glucocorticoids are directly related to the potency ranking and reduces requirements for corticosteroid therapy.86
of the compound and the length of use, so the clinician The approval of topical calcineurin inhibitors for
needs to instruct the patients and to balance the need for a treatment of atopic dermatitis is a great advance in our
more potent steroid with the potential for side-effects. management options for this disease. However, we should
Therefore, ultra-high-potency glucocorticoids should be still develop guidelines for use of topical corticosteroids
used only for very short periods of time and in regions that versus calcineurin inhibitors. Topical calcineurin
are lichenified but not on the face or intertriginous areas. inhibitors can be advantageous over topical corticos-
Tacrolimus—Topically applied FK-506, or tacrolimus, a teroids in some circumstances, including treatment of
calcineurin inhibitor that acts by binding with high affinity patients who do not respond well to topical steroids,
to the 12 kDa macrophilin, has been successfully used in patients with steroid phobia, and treatment of face and
treatment of atopic dermatitis. Tacrolimus inhibits neck dermatitis where ineffective, low-potency topical
activation of several key cells involved in atopic dermatitis, corticosteroids are usually used because of fears of
including T cells, dendritic cells, mast cells, and steroid-induced skin atrophy. The potential use of topical
keratinocytes.77 Multicentre blinded vehicle-controlled calcineurin inhibitors as maintenance therapy is also
phase-3 trials with 0·03% and 0·1% tacrolimus ointment, intriguing for prevention of atopic dermatitis flares.86
have shown tacrolimus to be both effective and safe in However, although systemic absorption of these
adults and children.78–80 Local burning sensation has been compounds are low, the drugs need to be carefully
the only common adverse event and most patients have monitored to rule out the possibility that skin cancers and
greatly reduced pruritus within 3 days of starting increased viral skin infections will appear when such
treatment. In adults, but not children, a dose-response agents are used long-term.
effect was seen between 0·03% and 0·1% tacrolimus, Tar preparations—Coal tar preparations have anti-
especially for patients with severe skin disease. pruritic and anti-inflammatory effects on the skin but are
Long-term open-label studies81 with tacrolimus less active than glucocorticoids. Their use is restricted to
ointment applied on up to 100% body surface area have chronic lesions and they can induce folliculitis and photo-
been done in adults and children with sustained efficacy sensitivity. There is a theoretical risk of tar being a
and no important side-effects—eg, no increased skin carcinogen based on observational studies of workers
infections. In addition, unlike topical glucocorticoids, using tar components in their occupations.
tacrolimus ointment is not atrophogenic and has been
used safely for facial and eyelid eczema. Of note, three Anti-infectious management
children with Netherton syndrome, who responded to Antistaphylococcal antibiotics can be helpful in treatment
treatment with 0·1% tacrolimus ointment, had of poorly-controlled patients who are heavily colonised or
substantial percutaneous absorption of the drug, with infected with S aureus.87 Topical mupirocin or fusidic acid
serum concentrations well above the therapeutic range.82 is useful in treatment of localised impetiginised lesions. In
In retrospect, this observation is not surprising, since patients with extensive superinfection with sensitive
patients with Netherton syndrome have a skin barrier S aureus strains, a course of systemic antibiotics such as
dysfunction that puts them at risk for increased erythromycin and the newer macrolide antibiotics
percutaneous absorption of topical drugs. Since patients (azithromycin and clarithromycin) is usually beneficial.
with Netherton syndrome are sometimes misdiagnosed However, for macrolide-resistant S aureus, a penicillinase-
as having atopic dermatitis, this diagnosis should be resistant penicillin (dicloxacillin, oxacillin, or cloxacillin)
considered in any infant or child with extensive and first-generation cephalosporins might be preferred.
erythroderma resistant to treatment and monitored for Atopic dermatitis can be complicated by recurrent viral
blood tacrolimus concentrations as deemed appropriate skin infections such as warts and mollusca contagiosa
by the care provider. Increased blood concentrations of which can reflect local defects in T-cell function.88 Herpes
tacrolimus have generally not been reported in patients simplex, resulting in Kaposi’s varicelliform eruption
with moderate to severe atopic dermatitis. Indeed, or eczema herpeticum can be a serious infection. After
quantifiable tacrolimus blood concentrations are isolated an incubation period of 5–12 days, many itchy
and transient occurrences seen only early in treatment. vesiculopustular lesions erupt in a disseminated pattern;
However, the potential risk of systemic absorption must vesicular lesions are umbilicated, tend to crop, and
be considered in any child with extensive skin disease often become haemorrhagic and crusted (figure 4). The
because of their high ratio of body surface area to presence of punched-out erosions, vesicles, or infected
weight. In this situation, especially if patients develop skin lesions that do not respond to oral antibiotics should
infection or changes in liver or kidney function, plasma initiate a search for herpes simplex. Antiviral treatment for
drug concentrations should be obtained. 0·03% cutaneous herpes simplex infections is of crucial
tacrolimus ointment has been approved for short-term importance in patients with widespread atopic dermatitis
and intermittent long-term use in for children since life-threatening dissemination has been reported. In
2–15 years of age with moderate to severe atopic patients with atopic dermatitis, smallpox vaccination or
dermatitis; and 0·03% and 0·1% for adults. even exposure to vaccinated individuals can cause a severe

156 THE LANCET • Vol 361 • January 11, 2003 •

For personal use. Only reproduce with permission from The Lancet Publishing Group.

been reported in a few patients with severe, resistant

atopic dermatitis who were given extracorporeal
photopheresis and topical glucocorticoids.

Systemic treatment
Antihistamines—Reduction of skin inflammation and
dryness with topical glucocorticoids and skin care,
respectively, will often symptomatically reduce pruritus.
Systemic antihistamines act mainly by blocking the H1
receptors on dermal mast cells. Since histamine is only
one of many mediators that can induce pruritus, some
patients might derive minimum benefit from treatment
with antihistamines. Results of studies92 of the
effectiveness of newer non-sedating antihistamines in
control of pruritus have varied greatly, although these
drugs could be useful in patients with atopic dermatitis
and concomitant urticaria.
Since pruritus is usually worse at night than during the
day, sedating antihistamines such as hydroxyzine or
diphenhydramine offer an advantage when used at
bedtime. Doxepin hydrochloride has both tricyclic
antidepressant and H1-histamine and H2-histamine
receptor blocking effects. If nocturnal pruritus remains
severe, short-term use of a sedative to allow adequate rest
might be appropriate. Treatment of atopic dermatitis
with topical antihistamines is useless and can induce
cutaneous sensitisation.
Systemic glucocorticoids—Use of systemic gluco-
Figure 4: Eczema herpeticum corticoids, such as oral prednisone, is rarely indicated,
Magnification ⫻4. since the clinical improvement is frequently followed by a
severe rebound flare at discontinuation. Short courses of
widespread skin rash called eczema vaccinatum similar in oral glucocorticoids are sometimes appropriate for an
appearance to eczema herpeticum. acute exacerbation while other treatment measures are
Patients with atopic dermatitis have increased being instituted. In this case, it is important to taper the
prevalence of fungal infections compared with non-atopic dose and begin intensified skin care, especially with
controls. The role of Malassezia furfur (Pityrosporum topical glucocorticoids and bathing followed by
orbiculare) in atopic dermatitis has been of particular application of emollients, to prevent rebound flares.
interest. M furfur is a lipophilic yeast commonly present in Interferon gamma—Interferon gamma downregulates
the seborrheic regions of the skin and in the scalp. IgE Th2 cell function. Treatment with recombinant
antibodies against this organism are frequently found in interferon gamma results in clinical improvement.93
patients with atopic dermatitis and most frequently in Reduction in clinical severity of atopic dermatitis
patients with head and neck dermatitis.89 The potential correlated with the ability of this drug to decrease blood
importance of M furfur and other dermatophyte infections eosinophilia. However, influenza-like symptoms are
is further supported by the reduction of atopic dermatitis commonly seen side-effects early in the treatment course
skin severity in such patients after treatment with and limit use of this treatment.
antifungal agents. Ciclosporin—Ciclosporin is a potent immuno-
suppressive drug that acts mostly on T cells by
Phototherapy suppressing cytokine transcription. The drug binds to
Natural sunlight is frequently beneficial to patients with intracellular cyclophilin and thereby acts in a similar way
atopic dermatitis. However, if the sunlight occurs to tacrolimus and pimecrolimus. Children and adults
together with high heat or humidity, thereby triggering with severe atopic dermatitis refractory to conventional
sweating and pruritus, it can be deleterious to patients. treatment, can benefit from short-term ciclosporin, with
Broad-band ultraviolet B, broad-band ultraviolet A, improved quality of life.94 However, discontinuation of
narrow-band ultraviolet B (311 nm), ultraviolet A-1 treatment generally results in relapse flares. Raised serum
(340–400 nm), and combined ultraviolet A and B creatinine or more substantial renal impairment and
phototherapy can be useful adjuncts in treatment of hypertension are specific side-effects of concern.
atopic dermatitis.90,91 Photochemotherapy with psoralen Antimetabolites—Mycophenolate mofetil is a purine
and ultraviolet A (PUVA) should be restricted to patients biosynthesis inhibitor used as an immunosuppressant in
with severe, widespread atopic dermatitis, although few organ transplantation, has been used for treatment of
investigators have compared this treatment with other refractory inflammatory skin disorders.95 Short-term oral
modes of phototherapy. Short-term adverse effects with mycophenolate mofetil 2 g daily as monotherapy results
phototherapy can include erythema, skin pain, pruritus, in clearing of skin lesions in adults with atopic dermatitis
and pigmentation. Potential long-term adverse effects resistant to other treatment including topical and oral
include premature skin aging and cutaneous malignant steroids and PUVA. The drug is generally well tolerated,
diseases. with the exception of occasional herpes retinitis. Dose-
Extracorporeal photopheresis consists of the passage of related bone marrow suppression has also been recorded.
psoralen-treated leukocytes through an extracorporeal Not all patients benefit from treatment. Therefore, the
ultraviolet A light system. Clinical improvement in skin drug should be discontinued if patients do not respond
lesions associated with reduced IgE concentrations have within 4–8 weeks.

THE LANCET • Vol 361 • January 11, 2003 • 157

For personal use. Only reproduce with permission from The Lancet Publishing Group.

Methotrexate is an antimetabolite with potent inhibitory 6 Schäfer T, Krämer U, Vieluf D, Abeck D, Behrendt H, Ring J. The
effects on inflammatory cytokine synthesis and cell excess of atopic eczema in East Germany is related to the intrinsic
chemotaxis. It has been used for patients with recalcitrant type. Br J Dermatol 2000; 143: 992–98.
7 von Mutius E. The environmental predictors of allergic disease.
atopic dermatitis, though controlled trials are few. J Allergy Clin Immunol 2000; 105: 9–19.
Azathioprine is a purine analogue with anti- 8 Romagnani S. The role of lymphocytes in allergic disease.
inflammatory and antiproliferative effects, which has J Allergy Clin Immunol 2000; 105: 399–408.
been used for severe atopic dermatitis, though no 9 Williams HC. Diagnostic criteria for atopic dermatitis: where do we
controlled trials have been reported. Myelosuppression is go from here? Arch Dermatol 1999; 135: 583–86.
an important adverse effect, and thiopurinemethyl 10 Leung DY. Atopic dermatitis: new insights and opportunities for
therapeutic intervention. J Allergy Clin Immunol 2000; 105: 860–76.
transferase concentrations can predict individuals at risk. 11 Saurat JH. Eczema in primary immune-deficiencies: clues to the
pathogenesis of atopic dermatitis with special reference to the
Complementary approaches Wiskott-Aldrich syndrome. Acta Derm Venereol Suppl 1985; 114:
Emotional stressors—Patients with atopic dermatitis often 125–28.
respond to frustration, embarrassment, or other stressful 12 Woodward AL, Spergel JM, Alenius H, et al. An obligate role for
T-cell receptor alphabeta+ T cells but not T-cell receptor
events with increased pruritus and scratching, and stress gammadelta+ T cells, B cells, or CD40/CD40L interactions in a
can induce immunological changes in such patients.96,97 mouse model of atopic dermatitis. J Allergy Clin Immunol 2001; 107:
Psychological assessment or counselling should be 359–66.
considered in patients who have difficulty with emotional 13 Beck LA, Leung DY. Allergen sensitization through the skin induces
systemic allergic responses. J Allergy Clin Immunol 2000; 106 (suppl):
triggers or psychological problems contributing to S258–63.
difficulty in managing their disease. Relaxation, 14 Higashi N, Gesser B, Kawana S, Thestrup-Pedersen K. Expression of
behavioural modification, or biofeedback can be helpful in IL-18 mRNA and secretion of IL-18 are reduced in monocytes from
patients with habitual scratching. patients with atopic dermatitis. J Allergy Clin Immunol 2001; 108:
Other treatments—Leucocytes from patients with 607–14.
15 van Reijsen FC, Bruijnzeel-Koomen CA, Kalthoff FS, et al. Skin-
atopic dermatitis have increased cAMP-PDE enzyme derived aeroallergen-specific T-cell clones of Th2 phenotype in
activity. Monocytes from such patients produce raised patients with atopic dermatitis. J Allergy Clin Immunol 1992; 90:
concentrations of PGE2 and interleukin 10, which both 184–93.
inhibit production of interferon ␥ by T cells. Topical 16 Spergel JM, Mizoguchi E, Brewer JP, Martin TR, Bhan AK, Geha
application of PDE inhibitors has shown clinical benefit in RS. Epicutaneous sensitization with protein antigen induces localized
allergic dermatitis and hyperresponsiveness to methacholine after
atopic dermatitis.17 Intravenous immunoglobulin reduces single exposure to aerosolized antigen in mice. J Clin Invest 1998;
skin inflammation in patients with refractory atopic 101: 1614–22.
dermatitis.98 Results of several studies have also suggested 17 Hanifin JM, Chan SC, Cheng JB, et al. Type 4 phosphodiesterase
that patients with atopic dermatitis benefit from treatment inhibitors have clinical and in vitro anti-inflammatory effects in atopic
dermatitis. J Invest Dermatol 1996; 107: 51–56.
with traditional Chinese herbal therapy.99 The possibility
18 Jirapongsananuruk O, Hofer MF, Trumble AE, Norris DA,
of hepatic toxic effects, cardiac side-effects, or Leung DY. Enhanced expression of B7.2 (CD86) in patients with
idiosyncratic reactions, however, is a concern. The atopic dermatitis: a potential role in the modulation of IgE synthesis.
specific ingredients of the herbs also remain to be J Immunol 1998; 160: 4622–27.
elucidated and some preparations have been found to be 19 Mihm MC Jr, Soter NA, Dvorak HF, Austen KF. The structure of
normal skin and the morphology of atopic eczema. J Invest Dermatol
contaminated with glucocorticoids.100 1976; 67: 305–12.
20 Leung DY, Bhan AK, Schneeberger EE, Geha RS. Characterization
Future directions of the mononuclear cell infiltrate in atopic dermatitis using
Atopic dermatitis is often the first presentation of an monoclonal antibodies. J Allergy Clin Immunol 1983; 71: 47–56.
individual destined to a lifetime of allergy and asthma. 21 Wollenberg A, Kraft S, Hanau D, Bieber T. Immunomorphological
and ultrastructural characterization of Langerhans cells and a novel,
Since the skin is a highly sensitising organ that contributes inflammatory dendritic epidermal cell (IDEC) population in lesional
greatly to the systemic allergic response, highly effective skin of atopic eczema. J Invest Dermatol 1996; 106: 446–53.
treatments need to be developed to reduce skin 22 Hamid Q, Boguniewicz M, Leung DY. Differential in situ cytokine
inflammation in this disease. Advances are likely to need gene expression in acute versus chronic atopic dermatitis. J Clin
better definitions for the various clinical phenotypes of Invest 1994; 94: 870–76.
23 Hamid Q, Naseer T, Minshall EM, Song YL, Boguniewicz M,
atopic dermatitis, including identification of the genes Leung DY. In vivo expression of IL-12 and IL-13 in atopic
leading to the disease, a better understanding of the dermatitis. J Allergy Clin Immunol 1996; 98: 225–31.
immunoregulatory abnormalities underlying it, and new 24 Grewe M, Bruijnzeel-Koomen CA, Schopf E, et al. A role for Th1
paradigms for preventing relapses of this skin disorder. and Th2 cells in the immunopathogenesis of atopic dermatitis.
Such advances will probably be tied to development of Immunol Today 1998; 19: 359–61.
25 Trautmann A, Akdis M, Schmid-Grendelmeier P, et al. Targeting
pharmacogenetics and targeting of effective treatments to keratinocyte apoptosis in the treatment of atopic dermatitis and
subsets of patients with atopic dermatitis. allergic contact dermatitis. J Allergy Clin Immunol 2001; 108: 839–46.
26 Novak N, Kraft S, Bieber T. IgE receptors. Curr Opin Immunol 2001;
13: 721–26.
References 27 von Bubnoff D, Geiger E, Bieber T. Antigen-presenting cells in
allergy. J Allergy Clin Immunol 2001; 108: 329–39.
1 Bieber T, Leung DYM, eds. Atopic dermatitis. New York: Marcel 28 Langeveld-Wildschut EG, Bruijnzeel PL, et al. Clinical and
Dekker, 2002. immunologic variables in skin of patients with atopic eczema and
2 Johansson SG, Hourihane JO, Bousquet J, et al. A revised either positive or negative atopy patch test reactions.
nomenclature for allergy: an EAACI position statement from the J Allergy Clin Immunol 2000; 105: 1008–16.
EAACI nomenclature task force. Allergy 2001; 56: 813–24. 29 Laberge S, Ghaffar O, Boguniewicz M, Center DM, Leung DY,
3 Schultz-Larsen F, Hanifin JM. Epidemiology of atopic dermatitis. Hamid Q. Association of increased CD4+ T-cell infiltration with
Immunol Allergy Clin North Am 2002; 22: 1–24. increased IL-16 gene expression in atopic dermatitis.
4 Taylor B, Wadsworth J, Wadsworth M, Peckham C. Changes in the J Allergy Clin Immunol 1998; 102: 645–50.
reported prevalence of childhood eczema since the 1939–45 war. 30 Reich K, Hugo S, Middel P, Blaschke V, et al. Evidence for a role of
Lancet 1984; 2: 1255–57. Langerhans cell-derived IL-16 in atopic dermatitis.
5 Williams H, Robertson C, Stewart A, et al. Worldwide variations in J Allergy Clin Immunol 2002; 109: 681–87.
the prevalence of symptoms of atopic eczema in the International 31 Taha RA, Minshall EM, Leung DY, et al. Evidence for increased
Study of Asthma and Allergies in Childhood. J Allergy Clin Immunol expression of eotaxin and monocyte chemotactic protein-4 in atopic
1999; 103: 125–38. dermatitis. J Allergy Clin Immunol 2000; 105: 1002–07.

158 THE LANCET • Vol 361 • January 11, 2003 •

For personal use. Only reproduce with permission from The Lancet Publishing Group.

32 Yawalkar N, Uguccioni M, Scharer J, et al. Enhanced expression of 57 Valenta R, Seiberler S, Natter S, et al. Autoallergy: a pathogenetic
eotaxin and CCR3 in atopic dermatitis. J Invest Dermatol 1999; 113: factor in atopic dermatitis? J Allergy Clin Immunol 2000; 105: 432–37.
43–48. 58 Valenta R, Natter S, Seiberler S, et al. Molecular characterization of
33 Morales J, Homey B, Vicari AP, et al. CTACK, a skin-associated an autoallergen, Hom s 1, identified by serum IgE from atopic
chemokine that preferentially attracts skin-homing memory T cells. dermatitis patients. J Invest Dermatol 1998; 111: 1178–83.
Proc Natl Acad Sci USA 1999; 96: 14470–75. 59 Kinaciyan T, Natter S, Kraft D, Stingl G, Valenta R. IgE
34 Galli G, Chantry D, Annunziato F, et al. Macrophage-derived autoantibodies monitored in a patient with atopic dermatitis under
chemokine production by activated human T cells in vitro and in cyclosporin A treatment reflect tissue damage. J Allergy Clin Immunol
vivo: preferential association with the production of type 2 cytokines. 2002; 109: 717–19.
Eur J Immunol 2000; 30: 204–10. 60 Leyden JJ, Kligman AM. The case for steroid-antibiotic
35 Kakinuma T, Nakamura K, Wakugawa M, et al. Thymus and combinations. Br J Dermatol 1977; 96: 179–87.
activation-regulated chemokine in atopic dermatitis: serum thymus 61 Leung DY, Harbeck R, Bina P, et al. Presence of IgE antibodies to
and activation-regulated chemokine level is closely related with staphylococcal exotoxins on the skin of patients with atopic dermatitis.
disease activity. J Allergy Clin Immunol 2001; 107: 535–41. Evidence for a new group of allergens. J Clin Invest 1993; 92: 1374–80.
36 Nakatani T, Kaburagi Y, Shimada Y, et al. CCR4 memory CD4+ 62 Breuer K, Wittmann M, Bosche B, Kapp A, Werfel T. Severe atopic
T lymphocytes are increased in peripheral blood and lesional skin dermatitis is associated with sensitization to staphylococcal
from patients with atopic dermatitis. J Allergy Clin Immunol 2001; enterotoxin B (SEB). Allergy 2000; 55: 551–55.
107: 353–58. 63 Bunikowski R, Mielke ME, Skarabis H, et al. Evidence for a disease-
37 Bratton DL, Hamid Q, Boguniewicz M, Doherty DE, Kailey JM, promoting effect of Staphylococcus aureus-derived exotoxins in atopic
Leung DY. Granulocyte macrophage colony-stimulating factor dermatitis. J Allergy Clin Immunol 2000; 105: 814–19.
contributes to enhanced monocyte survival in chronic atopic 64 Strickland I, Hauk PJ, Trumble AE, Picker LJ, Leung DY. Evidence
dermatitis. J Clin Invest 1995; 95: 211–18. for superantigen involvement in skin homing of T cells in atopic
38 Nickoloff BJ, Naidu Y. Perturbation of epidermal barrier function dermatitis. J Invest Dermatol 1999; 112: 249–53.
correlates with initiation of cytokine cascade in human skin. 65 Hofer MF, Harbeck RJ, Schlievert PM, Leung DY. Staphylococcal
J Am Acad Dermatol 1994; 30: 535–46. toxins augment specific IgE responses by atopic patients exposed to
39 Giustizieri ML, Mascia F, Frezzolini A, et al. Keratinocytes from allergen. J Invest Dermatol 1999; 112: 171–76.
patients with atopic dermatitis and psoriasis show a distinct 66 Hauk PJ, Hamid QA, Chrousos GP, Leung DY. Induction of
chemokine production profile in response to T cell-derived cytokines. corticosteroid insensitivity in human PBMCs by microbial
J Allergy Clin Immunol 2001; 107: 871–77. superantigens. J Allergy Clin Immunol 2000; 105: 782–87.
40 Soumelis V, Reche PA, Kanzler H, et al. Human epithelial cells 67 Skov L, Olsen JV, Giorno R, Schlievert PM, Baadsgaard O,
trigger dendritic cell mediated allergic inflammation by producing Leung DY. Application of Staphylococcal enterotoxin B on normal
TSLP. Nat Immunol 2002; 3: 673–80. and atopic skin induces up-regulation of T cells by a superantigen-
41 Forrest S, Dunn K, Elliott K, et al. Identifying genes predisposing to mediated mechanism. J Allergy Clin Immunol 2000; 105: 820–26.
atopic eczema. J Allergy Clin Immunol 1999; 104: 1066–70. 68 Nilsson EJ, Henning CG, Magnusson J. Topical corticosteroids and
42 Kawashima T, Noguchi E, Arinami T, et al. Linkage and association Staphylococcus aureus in atopic dermatitis. J Am Acad Dermatol 1992;
of an interleukin 4 gene polymorphism with atopic dermatitis in 27: 29–34.
Japanese famillies. J Med Genet 1998; 35: 502–04. 69 Remitz A, Kyllonen H, Granlund H, Reitamo S. Tacrolimus
43 Cox HE, Moffatt MF, Faux JA, et al. Association of atopic dermatitis ointment reduces staphylococcal colonization of atopic dermatitis
to the beta subunit of the high affinity immunoglobulin E receptor. lesions. J Allergy Clin Immunol 2001; 107: 196–97.
Br J Dermatol 1998; 138: 182–87. 70 Cho SH, Strickland I, Tomkinson A, Fehringer AP, Gelfand EW,
44 Arkwright PD, Chase JM, Babbage S, Pravica V, David TJ, Leung DY. Preferential binding of Staphylococcus aureus to skin sites
Hutchinson IV. Atopic dermatitis is associated with a low-producer of Th2-mediated inflammation in a murine model. J Invest Dermatol
transforming growth factor beta(1) cytokine genotype. 2001; 116: 658–63.
J Allergy Clin Immunol 2001; 108: 281–84. 71 Cho SH, Strickland I, Boguniewicz M, Leung DY. Fibronectin and
45 Lee YA, Wahn U, Kehrt R, et al. A major susceptibility locus for fibrinogen contribute to the enhanced binding of Staphylococcus
atopic dermatitis maps to chromosome 3q21. Nat Genet 2000; 26: aureus to atopic skin. J Allergy Clin Immunol 2001; 108: 269–74.
470–73. 72 Ong PY, Ohtake T, Brandt C, et al. Decreased antimicrobial
46 Cookson WO, Ubhi B, Lawrence R, et al. Genetic linkage of peptides contribute to increased infections in atopic dermatitis.
childhood atopic dermatitis to psoriasis susceptibility loci. Nat Genet N Engl J Med 2002; 347: 1151–60.
2001; 27: 372–73. 73 Hoare C, Li Wan Po A, Williams H. Systematic review of treatments
47 Sampson HA. Food allergy. Part 1: immunopathogenesis and clinical for atopic eczema. Health Technol Assess 2000; 4: 1–191.
disorders. J Allergy Clin Immunol 1999; 103: 717–28. 74 Imokawa G. Lipid abnormalities in atopic dermatitis.
48 Lever R, MacDonald C, Waugh P, Aitchison T. Randomised J Am Acad Dermatol 2001; 45 (suppl): S29–32.
controlled trial of advice on an egg exclusion diet in young children 75 Woodmansee DP, Christiansen SC. Improvement in atopic
with atopic eczema and sensitivity to eggs. Pediatr Allergy Immunol dermatitis in infants with the introduction of an elemental formula.
1998; 9: 13–19. J Allergy Clin Immunol 2001; 108: 309.
49 van Reijsen FC, Felius A, Wauters EA, Bruijnzeel-Koomen CA, 76 Van Der Meer JB, Glazenburg EJ, Mulder PG, Eggink HF,
Koppelman SJ. T-cell reactivity for a peanut-derived epitope in the Coenraads PJ, for the Netherlands Adult Atopic DermatitisStudy
skin of a young infant with atopic dermatitis. J Allergy Clin Immunol Group. The management of moderate to severe atopic dermatitis in
1998; 101: 207–09. adults with topical fluticasone propionate. Br J Dermatol 1999; 140:
50 Li XM, Kleiner G, Huang CK, et al. Murine model of atopic 1114–21.
dermatitis associated with food hypersensitivity. 77 Wollenberg A, Sharma S, von Bubnoff D, Geiger E, Haberstok J,
J Allergy Clin Immunol 2001; 107: 693–702. Bieber T. Topical tacrolimus (FK506) leads to profound phenotypic
51 Tupker RA, De Monchy JG, Coenraads PJ, Homan A, and functional alterations of epidermal antigen-presenting dendritic
van der Meer JB. Induction of atopic dermatitis by inhalation cells in atopic dermatitis. J Allergy Clin Immunol 2001; 107: 519–25.
of house dust mite. J Allergy Clin Immunol 1996; 97: 1064–70. 78 Reitamo S, Van Leent EJ, Ho V, et al. Efficacy and safety of
52 Wheatley LM, Platts-Mills TAE. Role of inhalant allergens in atopic tacrolimus ointment compared with that of hydrocortisone acetate
dermatitis. In: Leung DYM, Greaves MW, eds. Allergic skin disease: ointment in children with atopic dermatitis. J Allergy Clin Immunol
a multidisciplinary approach. New York: Marcel Dekker, 2000: 423. 2002; 109: 539–46.
53 Tan BB, Weald D, Strickland I, Friedmann PS. Double-blind 79 Reitamo S, Rustin M, Ruzicka T, et al. Efficacy and safety of
controlled trial of effect of housedust-mite allergen avoidance on tacrolimus ointment compared with that of hydrocortisone butyrate
atopic dermatitis. Lancet 1996; 347: 15–18. ointment in adult patients with atopic dermatitis. J Allergy Clin
54 Holm L, Bengtsson A, van Hage-Hamsten M, Ohman S, Immunol 2002; 109: 547–55.
Scheynius A. Effectiveness of occlusive bedding in the treatment 80 Paller A, Eichenfield LF, Leung DY, Stewart D, Appell M. A
of atopic dermatitis—a placebo-controlled trial of 12 months’ 12-week study of tacrolimus ointment for the treatment of atopic
duration. Allergy 2001; 56: 152–58. dermatitis in pediatric patients. J Am Acad Dermatol 2001;
55 Scalabrin DM, Bavbek S, Perzanowski MS, Wilson BB, 44 (suppl): S47–57.
Platts-Mills TA, Wheatley LM. Use of specific IgE in assessing the 81 Kang S, Lucky AW, Parisen D, et al. Long-term safety and efficacy
relevance of fungal and dust mite allergens to atopic dermatitis: of tacrolimus ointment for the treatment of atopic dermatitis in
a comparison with asthmatic and nonasthmatic control subjects. children. J Am Acad Dermatol 2001; 44 (suppl): S58–64.
J Allergy Clin Immunol 1999; 104: 1273–79. 82 Allen A, Siegfried E, Silverman R, et al. Significant absorption of
56 Keller P. Beitrag zu den Beziehungen von Asthma und Ekzem. topical tacrolimus in 3 patients with Netherton syndrome.
Arch Derm Syph Berl 1924; 148: 82. Arch Dermatol 2001; 137: 747–50.

THE LANCET • Vol 361 • January 11, 2003 • 159

For personal use. Only reproduce with permission from The Lancet Publishing Group.

83 Zuberbier T, Chong SU, Grunow K, et al. The ascomycin adult atopic eczema: a randomised controlled trial. Lancet 2001; 357:
macrolactam pimecrolimus (Elidel, SDZ ASM 981) is a potent 2012–16.
inhibitor of mediator release from human dermal mast cells and 92 Simons FE. Prevention of acute urticaria in young children with
peripheral blood basophils. J Allergy Clin Immunol 2001; 108: 275–80. atopic dermatitis. J Allergy Clin Immunol 2001; 107: 703–06.
84 Luger T, Van Leent EJ, Graeber M, et al. SDZ ASM 981: an 93 Hanifin JM, Schneider LC, Leung DY, et al. Recombinant
emerging safe and effective treatment for atopic dermatitis. interferon gamma therapy for atopic dermatitis. J Am Acad Dermatol
Br J Dermatol 2001; 144: 788–94. 1993; 28: 189–97.
85 Eichenfield LF, Lucky AW, Boguniewicz M, et al. Safety and efficacy 94 Sowden JM, Berth-Jones J, Ross JS, et al. Double-blind, controlled,
of pimecrolimus (ASM 981) cream 1% in the treatment of mild and crossover study of cyclosporin in adults with severe refractory atopic
moderate atopic dermatitis in children and adolescents. dermatitis. Lancet 1991; 338: 137–40.
J Am Acad Dermatol 2002; 46: 495–504. 95 Grundmann-Kollmann M, Podda M, Ochsendorf F,
86 Kapp, A, Papp K, Bingham A, et al. Long-term management of Boehncke WH, Kaufmann R, Zollner TM. Mycophenolate mofetil is
atopic dermatitis in infants with topical pimecrolimus, a non-steroid effective in the treatment of atopic dermatitis. Arch Dermatol 2001;
anti-inflammatory drug. J Allergy Clin Immunol 2002; 110: 277–84. 137: 870–73.
87 Boguniewicz M, Sampson H, Leung SB, Harbeck R, Leung DY. 96 Schmid-Ott G, Jaeger B, Meyer S, Stephan E, Kapp A, Werfel T.
Effects of cefuroxime axetil on Staphylococcus aureus colonization and Different expression of cytokine and membrane molecules by
superantigen production in atopic dermatitis. J Allergy Clin Immunol circulating lymphocytes on acute mental stress in patients with
2001; 108: 651–52. atopic dermatitis in comparison with healthy controls.
88 Ambach A, Bonnekoh B, Gollnick H. Perforin hyperreleasability J Allergy Clin Immunol 2001; 108: 455–62.
and depletion in cytotoxic T cells from patients with exacerbated 97 Schmid-Ott G, Jaeger B, Adamek C, et al. Levels of circulating
atopic dermatitis and asymptomatic rhinoconjunctivitis allergica. CD8(+) T lymphocytes, natural killer cells, and eosinophils increase
J Allergy Clin Immunol 2001; 107: 878–86. upon acute psychosocial stress in patients with atopic dermatitis.
89 Zargari A, Eshaghi H, Back O, Johansson S, Scheynius A. Serum IgE J Allergy Clin Immunol 2001; 107: 171–77.
reactivity to Malassezia furfur extract and recombinant M furfur 98 Jolles S. A review of high-dose intravenous immunoglobulin
allergens in patients with atopic dermatitis. Acta Derm Venereol 2001; treatment for atopic dermatitis. Clin Exp Dermatol 2002;
81: 418–22. 27: 3–7.
90 Krutmann J, Diepgen TL, Luger TA, et al. High-dose UVA1 therapy 99 Koo J, Arain S. Traditional Chinese medicine for the treatment of
for atopic dermatitis: results of a multicenter trial. dermatologic disorders. Arch Dermatol 1998; 134: 1388–93.
J Am Acad Dermatol 1998; 38: 589–93. 100 Keane FM, Munn SE, du Vivien AW, Taylor NF, Higgins EM.
91 Reynolds NJ, Franklin V, Gray JC, Diffey BL, Farr PM. Narrow- Analysis of Chinese herbal creams prescribed for dematological
band ultraviolet B and broad-band ultraviolet A phototherapy in conditions. BMJ 1999; 318: 563–64.

Uses of error
Laboratory diagnosis
Caroline Mackie Ogilvie

Scientists working in the area of laboratory diagnosis are in and these estimates are discussed with prospective
general further removed from the consequences of error patients. Our programme has been running successfully for
than clinicians who have direct contact with the patients. three years; however, we were shocked and saddened
But none of us likes to get things wrong, and diagnostic recently to hear that an affected baby had been born
errors in my own field of genetics, such as failing to identify following transfer of an embryo diagnosed as free from a
an abnormal chromosome on karyotype analysis, tend to serious genetic condition. The couple had declined
plunge one into gloom and self-reproach for several days, if prenatal diagnosis. Thorough evaluation of the test results
not weeks, depending on one’s personality. showed that the interpretation of the output was correct.
In the diagnostic laboratory, the final result of a test may Therefore the possibilities were that either there had been
be prone to error from various sources: inherent errors in an output error in the test, or that there had been
the test, leading to incorrect “output”; errors in the spontaneous natural conception. Statistical analysis
interpretation of the output (such as the example given revealed that a single abnormal pregnancy was at this point
above); and clerical errors in the transfer of information. well within our estimated error rate. Detailed analysis of
Inherent errors in the test are sometimes not appreciated; protocols revealed areas that could be improved—for
for instance, there is a considerable literature on the instance, better counselling of couples with respect to the
abnormal chromosome constitution of a large proportion use of contraception during their PGD cycle. Although the
of early human embryos in vitro, a conclusion based on potential error rate could be reduced further by
data obtained using a technique called FISH introducing extra parameters into the test, we realised that
(Fluorescence In Situ Hybridization). However, the any additional measurements to increase the sensitivity
inherent error rate using FISH is under-appreciated, and it would also decrease the specificity, thus increasing the
is possible that many of these so-called abnormalities may chance of excluding normal embryos from transfer due to
in fact have been due to FISH errors. Our department is errors in the new measurements. As many women
part of a successful Preimplantation Genetic Diagnosis undergoing PGD have small embryo cohorts, a decrease in
(PGD) centre, in which both FISH-based and PCR-based test specificity could significantly jeopardise their chances
tests are used to diagnose genetic abnormality in single of establishing a pregnancy. This case has given rise to
cells from human cleavage-stage embryos. These tests are useful discussions on the nature and value of error
technically very demanding and are prone to output errors estimation, and the best and clearest way of presenting
such as those caused by signal overlap (FISH) or single these complex ideas to our patients. We hope that
tube contamination (PCR). The nature of such single cell knowledge of this error will encourage more couples to opt
work makes error estimations difficult, but we do our best, for prenatal diagnosis following PGD.

Genetics Centre, Guy’s and St Thomas’ Hospital, London SE1 9RT, UK (C M Ogilvie DPhil)

160 THE LANCET • Vol 361 • January 11, 2003 •

For personal use. Only reproduce with permission from The Lancet Publishing Group.