PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Atopic Eczema in Children: Management of Atopic Eczema in Children from Birth up to the Age of 12 Years. London: RCOG Press; 2007 Dec. (NICE Clinical Guidelines, No. 57.)

7Treatment

Many of the treatments available for atopic eczema have been used in children. In this chapter, the evidence for each treatment is considered, starting with the most simple and moving on to more complex treatment options.

7.1. Emollients

The skin provides a barrier to the loss of water and penetration of irritants and allergens from the environment. The skin’s outermost layer, the stratum corneum, provides the protective barrier, preventing water loss and controlling secretions via evaporation essential to keeping the skin’s elasticity and firmness. In atopic eczema this barrier is damaged, both in eczematous areas and in clinically unaffected skin.

Atopic eczema has a strong genetic component. Filaggrin is a protein that is very important for the strength of the corneocytes (brick-like components) of the skin barrier. Changes in the filaggrin gene have been identified in children with atopic eczema. These changes result in an abnormal form of the filaggrin protein, which means that the corneocytes will not be as strong as in a child who does not have atopic eczema.231 Interactions between genes responsible for the breakdown of the skin barrier and irritants such as soap and detergents can trigger flares of eczema.

Emollients (or moisturisers) act by occluding water loss from outer layers of the skin and by directly adding water to the dry outer layers of the skin, thereby providing a protective film over the skin to keep moisture in and irritants out. More than 30 different emollients and more than ten emollient bath additives are listed in the British National Formulary for Children (BNFC).232 Emollients are available in a variety of formulations (ointments, creams, lotions, gels and aerosol sprays). Ointments, such as white soft paraffin and liquid paraffin, are greasy in nature whereas creams and lotions contain water and are more acceptable cosmetically. Creams, lotions and gels contain preservatives to protect against microbial growth in the presence of water. Antiseptics added to emollients include triclosan, chlorhexidine hydrochloride and benzalkonium chloride. See Table 7.1 for descriptions of the uses of the various types of emollient product.

Table 7.1. Types of emollient product.

Table 7.1

Types of emollient product.

Studies considered in this section

A health technology assessment (HTA) of treatments for atopic eczema was checked for RCTs evaluating the use of emollients in children.26 Narrative reviews were also checked for studies of any design.233,234 Where available, controlled trials evaluating the effectiveness of emollients in children with atopic eczema were considered for this section. Where RCTs were not available, studies of any design were considered.

Overview of available evidence

One RCT evaluated the use of emollients for the treatment of atopic eczema in children.235 No clinical trials considered the quantity or frequency of use of emollients. No evidence was found for most of the emollients listed in the BNFC. Some evidence from studies of various designs were identified for aqueous cream,236 emollients containing urea or ceramide,68,237–239 an antimicrobial emollient,240 and bath emollient preparations.241–244 The steroid-sparing effect of emollients has also been considered in clinical studies.245–248 Studies evaluating emollients in conjunction with topical corticosteroid wet wrap therapies are considered in Section 7.4.249,250

Moisturiser containing oat extract and evening primrose oil

One RCT in children (n = 76, age 6 months to 12 years) compared SCORAD and CDLQI after 8 weeks’ twice-daily treatment with a moisturiser containing oat extract and evening primrose oil.235 The control group received no emollient, but both groups used a standard cleansing bar and topical corticosteroids were permitted. There was a significant reduction in CDLQI in the treatment group (P = 0.001) but not in the control group (P = 0.17). There was no significant reduction in SCORAD in either group. No between-group analyses of these outcome measures were reported. [EL = 1−]

Aqueous cream

An audit of children attending a paediatric dermatology clinic recorded the proportion of immediate cutaneous reactions to emollients (defined as one or more of burning, stinging, itching and redness developing within 20 minutes of application). Aqueous cream was the emollient used by most (71%), which was associated with an immediate cutaneous reaction in 56% of exposures, compared with 18% with other emollients used (details of the other emollients were not reported; n = 100).236 [EL = 3]

Preparations containing urea

Three studies evaluated preparations containing urea. None of the studies provide usable data for children with atopic eczema. One that compared urea 10% with betamethasone valerate 0.1% (a topical corticosteroid) in a within-patient (left–right side) trial in children with atopic eczema only reported the extent of improvement after 10 days’ treatment, providing no demographic data for the children nor numerical data for outcomes.237 Two other studies evaluating preparations containing urea were identified: in one of these it was not possible to tell whether any of the individuals treated were children with atopic eczema,238 and in the other no data were reported for the minority of children with atopic eczema.239

Ceramide-containing emollients

A within-patient (left–right side) comparison reported the use of a ceramide-containing emollient in addition to usual treatment for up to 20 weeks in children with atopic eczema (n = 24). The outcomes considered were severity (SCORAD) and skin parameters (transepidermal water loss, hydration, and integrity of the stratum corneum). However, results were only presented in graphs in the trial report, with no numerical data.68 [EL = 2−]

Bath emollients

Four studies considered the use of bath oil preparations; three provided some effectiveness data.241–244 Two studies which evaluated preparations containing antimicrobials241,243 are considered in Section 7.6.

A case series reported the use of a bath oil preparation containing soya oil plus lauromacrogols in children and young people with dry, itchy dermatoses (n = 3566). The diagnosis was atopic eczema in 86% of the cases, and most (94%) of those included were aged under 15 years. The bath oil was used daily by 13%, three times a week by 38%, twice a week by 42%, and once a week by 7%. Mean duration of treatment and follow-up was 6 weeks. Overall, 78% received other treatment for their skin condition, although details of these treatments were not reported. Therefore it is not known whether the improvements in the children’s global condition were due to the emollient or to other treatments. The study provided information on tolerability, with skin reactions reported in 0.28%. The reactions were described as mostly mild, and included burning, itching and reddening. Physician’s assessment of tolerability was ‘good’ in 97% of children.242 [EL = 3]

Frequency of bathing

The effects of using a bath emollient daily (by soaking one arm in a basin of water with added emollient) was evaluated in a within-patient (left–right side) comparison (n = 9). All children had standardised treatment consisting of weekly whole-body bathing in a bath containing the same emollient (Oilatum®), twice-daily application of an emollient and a topical corticosteroid, and use of emulsifying wax as soap substitute. The treated (daily treatment) and untreated (routine care) arms were evaluated by an assessor blind to treatment allocation. The mean difference in clinical score at 4 weeks (a measure of extent and severity of atopic eczema) was not significant, although the difference in the mean change in score over the duration of the 4 week study was reported to be significantly different.244 [EL = 2−]

Studies evaluating the steroid-sparing effect of emollients

Three controlled trials sought to evaluate the steroid-sparing effects of emollients.245,246,248 They all compared the use of an emollient plus a topical corticosteroid with a topical corticosteroid used alone.245,246 A lack of baseline data meant it was not known whether the groups were similar other than in the interventions made. [EL = 2−] Additionally, it was not clear in either study whether daily quantities of topical corticosteroids applied in the once-daily versus twice-daily groups were similar.

The first study was an RCT in infants (n = 162) comparing micronised desonide 0.1% (high potency) and/or desonide 0.1% (moderate potency) to the respective treatments plus an emollient containing evening primrose oil and oat extract.248 Emollient was applied twice daily to dry, non-inflamed areas of skin over the whole body in the treatment group and tubes of topical corticosteroid were weighed at 0, 3 and 6 weeks to assess the amount used by all participants. At 6 weeks, there was a significant difference between the treatment groups in the amount of high potency corticosteroid used (mean difference 6.14 g, P = 0.025). There were no significant differences in the amount of moderate potency topical corticosteroid used, SCORAD severity index or quality of life. Two participants experienced severe adverse effects and discontinued treatment. [EL = 1−]

The second study compared the effectiveness of hydrocortisone cream 2.5% applied twice daily with a regimen of hydrocortisone cream 2.5% plus an emollient, both applied once daily (n = 25). After 3 weeks’ treatment improvements in signs and symptoms of atopic eczema were reported in both groups, with no statistically significant difference between groups. However, there was poor reporting of outcomes.245 [EL = 2−]

The third study (n = 50) compared betamethasone valerate 0.1% applied twice daily with betamethasone valerate 0.1% applied in the morning and an emollient applied in the evening. After 4 weeks’ treatment there were no significant differences in improvements in SCORAD scores (P values were not stated). No adverse effects were reported during the trial.246 [EL = 2−]

A 1989 German trial compared the effects of fluprednidene 21 acetate (a topical corticosteroid; potency not reported) used twice daily without an emollient for 3 weeks with three other treatment regimens that involved using fluprednidene 21 acetate and its emollient base (n = 44). The three other groups were treated with the following:

  • fluprednidene 21 acetate on days 1 and 3 and emollient on day 2 (repeated until day 21)
  • fluprednidene 21 acetate on days 1 and 4, and emollient on days 2 and 3 (repeated until day 21)
  • fluprednidene 21 acetate on days 1 and 5 and emollient on days 2–4 (repeated until day 21).

The trial was published in German, but was summarised in an English language review paper.247 [EL = 3] It was not clear whether the patients were children or adults (or a mixture of the two). The study found that clinical outcomes (severity) were similar in the fluprednidene 21 acetate only group to the other three groups. The group using emollients for most days used 75% less fluprednidene 21 acetate than the group using the fluprednidene 21 acetate only. No numerical data were reported in the English language review paper.247

Cost-effectiveness

No cost-effectiveness studies were identified that addressed this clinical question.

Evidence statement for emollients

There was a lack of studies of any design that evaluated the effectiveness of emollients in children with atopic eczema. The available data consisted of isolated case series and case reports, with no controlled studies comparing emollients to placebo/no active intervention. With no control groups, it was not possible to quantify the benefits or harms of emollient therapy. Irritant adverse skin reactions such as stinging were documented to occur with emollients such as aqueous cream and bath oils. [EL = 3]

Case series that considered the effects of treatment with emollients containing antimicrobial agents (including bath oils) in children reported subjective global measures of improvement over the short term only (2–6 weeks). In these case series, children received other treatments and thus it was not possible to identify which treatment produced benefit. [EL = 3]

Although emollients are widely described as having a steroid-sparing effect, no robust data were identified to confirm or refute this. [EL = 2−]

From evidence to recommendations

The GDG believes that emollients are the most important treatment for atopic eczema because they restore the defective skin barrier. A complete emollient regimen produces optimum benefit. This involves avoidance of products that may irritate the skin or lead to breakdown of the skin barrier, including soaps, shampoo products and perfumed products obtained over the counter or on prescription. Adherence to an emollient regimen has the potential to reduce the need for more expensive treatments and associated GP consultations.

All children require an essential package of emollient therapy including a topical emollient and a wash product. A single emollient may satisfy both these functions. However, some children will require more than one product to ensure adequate emollient coverage. One of the most important environmental factors in triggering atopic eczema is soap and detergents. There are high levels of Staphylococcus aureus on the skin of children with atopic eczema (see Section 6.1) which are also an important trigger for flares of atopic eczema. Emollient bath oils and other emollient wash products provide an essential method to clean the skin without the damaging effect of soap and detergents.

Healthcare professionals should offer a range of different products to children with atopic eczema for topical application and for washing, and children should be encouraged to try out various combinations of topical products. The correct emollient is the one that the child will use.

Not all types of emollients suit all people. Adherence to emollient treatment is the key to successful therapy for atopic eczema. Children may have adverse reactions to some products, or may not like the way they feel on their skin. Topically applied emollients may be easier to apply on some children who can tolerate standing still for a period of time several times a day. Other children may need additional products that can be applied indirectly to the skin, such as in the bath, to ensure that adequate amounts of emollient are absorbed into their skin. Children’s preferences and tolerance for specific products will differ over time as they get older and their lifestyle and attitudes change. In addition, some bath products contain added ingredients (such as antimicrobials) that may be useful for short periods of time to manage specific conditions.

It is the GDG’s view that the practice of repeat prescribing of the same emollient products over long periods of time without review should be discouraged.

Idiosyncratic skin reactions/irritations and lifestyle may influence the choice of emollient. Since there is little cost difference between proprietary products, these factors should be taken into account when selecting an emollient in order to improve adherence to therapy. Although non-proprietary products are cheaper, they are often less acceptable to children and are not, therefore, usually suitable as a first-line treatment. Aqueous cream is associated with stinging when used as a leave-on emollient but can be used as a wash product. Since an emollient’s effectiveness and acceptability can change over time for a particular child, children and their parents/caregivers need to be encouraged to look for the signs that an emollient is no longer providing maximum benefit (for example, the return of symptoms of atopic eczema) and to seek the advice of a healthcare professional if they have concerns. They should then be offered an opportunity to try a different product or combination of products. [EL = 4]

Skin reactions, including stinging, are a manifestation of worsening eczema (breakdown of the skin barrier). Emollients are the mainstay of current treatment of atopic eczema, and clinical experience is that they reduce the need for topical corticosteroids. Regular use of emollients is essential to ensure rehydration of the skin, and to give skin flexibility. It is the experience of the GDG that dry skin requires a greasy emollient preparation, whereas red inflamed eczema usually responds better to water-based products because evaporation cools the skin. Other treatment for red inflamed eczema is discussed in Sections 7.2 and 7.7. [EL = 4]

The GDG’s view is that the effects of emollients are short-lived. Therefore, they should be used frequently and in large quantities, particularly after bathing or washing, in order to protect the integrity of the skin barrier. The frequency of use of the emollient will depend on the dryness of the child’s skin and the type of emollient used. It is the experience of the GDG that children with generalised atopic eczema typically require about 250 g per week or more of an emollient. This should far exceed the quantities of other treatments. [EL = 4]

It is the GDG’s view that the need for frequent application of emollients implies that children should have access to emollient therapy at nursery, pre-school or school. [EL = 4] The GDG noted that the selection of emollient preparations prescribed for each child could include conveniently sized containers for use outside the home (as well as large containers for use at home).

The GDG believes it is good practice to apply emollients by smoothing them into the skin in the direction of hair growth, rather than rubbing them in, to facilitate absorption and reduce the possibility of occlusion folliculitis. Rubbing introduces air which makes absorption more difficult. [EL = 4]

Recommendations for emollients (including research recommendations) are presented in Section 7.11.

7.2. Topical corticosteroids

Topical corticosteroids are derived from the naturally occurring corticosteroid cortisol (hydrocortisone) which is secreted by the adrenal cortex. Corticosteroids have anti-inflammatory and immunosuppressant effects, as well as other actions relevant to their effects on skin including inhibiting fibroblast proliferation and collagen synthesis, and local vasoconstriction.

Twenty topical corticosteroids are listed in the BNFC. They are available in a variety of formulations, including ointments, creams, and lotions. The available products also differ in potency (see Table 7.2). In the UK topical corticosteroids are divided into four categories: mild, moderate, potent and very potent. The potency of topical corticosteroids is usually determined by a vasoconstrictor assay that measures the degree and duration of blanching of the skin produced by topical application.251,252

Table 7.2. Potency of topical corticosteroids.

Table 7.2

Potency of topical corticosteroids.

The potency of a topical corticosteroid is not necessarily related to its concentration – it also depends on the specific modification (esterification) of the steroid molecule. For example, hydrocortisone (acetate) 1% is a mild preparation, but hydrocortisone butyrate 0.1% is a potent preparation. The clinical effect of a topical corticosteroid preparation depends on its potency, concentration and the formulation (vehicle/base).

The finger-tip unit253 is a validated method for applying topical corticosteroids in safe quantities. One finger-tip unit is a squeeze of cream or ointment along the index finger from the tip to the first finger joint. This weighs approximately half a gram and will cover a surface area of two adult hands (including the fingers). This information is often included in patient information leaflets.

Overview of available evidence

The HTA of treatments for atopic eczema was checked for evidence relating to children.26 Where available, RCTs evaluating the effectiveness of topical corticosteroids in children with atopic eczema were considered for this section. Where RCTs were not available, or were too short in duration to consider adverse effects, observational studies of any design were considered.

The NICE TA programme issued guidance on the frequency of application of topical corticosteroids in 2004.12 That guidance, which is adopted in this guideline, applies to both children and adults with atopic eczema. The HTA informing the NICE guidance included three studies involving children.12 No further RCTs considering frequency of application were identified.

Overall, ten RCTs compared topical corticosteroids of different potencies,254–263 four RCTs compared topical corticosteroids with other interventions (coal tar264 and topical calcineurin inhibitors265–267) and two RCTs compared different formulations of the same topical corticosteroid.268,269 Limited data comparing topical corticosteroids with placebo or no intervention in children only were found,270 and therefore studies that included both children and adults were also considered.271–273 Studies considering the steroid-sparing effects of emollients245,246,248 were described in Section 7.1.

Eight other reviews or studies of other design that considered only safety were also identified.274–281 One review of the safety of topical therapies for atopic eczema was identified, but no conclusions could be drawn in relation to children.282

No studies evaluating the use of the following topical corticosteroids in children with atopic eczema were identified: betamethasone valerate 0.025%, fludroxycortide (formerly known as flurandrenolone), fluocinolone acetonide 0.00625% or 0.025%, fluocortolone, beclometasone, betamethasone dipropionate 0.05%, diflucortolone 0.1% or 0.3%, fluocinonide 0.05% or clobetasol propionate 0.05%.

RCTs comparing topical corticosteroids with vehicle

One short-term (7 day) double-blind RCT reported the change in clinical score in children aged 4.5 months to 15 years with atopic eczema who were treated with desonide (a mild preparation) compared with its vehicle base (n = 40). The proportion showing improvement or clearance of their condition was significantly higher in the desonide group (67%) than in the vehicle group (16%, P < 0.001).270 [EL = 1+]

Other RCTs comparing a topical corticosteroid with placebo/vehicle included both children and adults, although none reported the proportion of children aged under 12 years and nor did they report data separately for this group.271–273 Each was a within-patient (left–right side) randomised double-blind comparison.

The first study compared hydrocortisone valerate 0.2% cream (moderately potent) with ‘placebo’ cream (no further details provided; n = 20).271 The creams were applied three times a day for 2 weeks. Although the study reported the proportion with clearance of the condition, no other details or numerical data were given. Clearance was reported for eight patients treated with hydrocortisone valerate 0.2% and in one treated with placebo.271 [EL = 1−]

The second study compared halcinonide 0.1% ointment (very potent), applied three times a day, with its vehicle base (n = 233, 92% completed and analysed). The global response was reported, though it was not clear exactly what was measured or how. The proportions with an excellent or good response were 85% and 44% in the halcinonide and placebo groups, respectively (P < 0.001).273 [EL = 1−]

The third study, in patients with mild to moderate atopic eczema, compared desonide 0.05% (mild potency) plus an emollient with desonide 0.05% alone (n = 80).272 After 3 weeks’ treatment, the reduction in severity score was significantly greater in the group treated with desonide plus emollient compared with desonide alone (80% versus 70%, respectively, P < 0.01). Global improvement of 75% or more was reported by 70% versus 55%, respectively (P < 0.01). Quantities of topical corticosteroid used were not reported. The proportions reporting burning or stinging on application during the first week of treatment were similar (12% versus 14%).272 [EL = 1+]

RCTs comparing different topical corticosteroids

Ten RCTs compared the effectiveness of topical corticosteroids of different potencies in children of various ages (2 months to 15 years), the majority including only children aged under 12 years with atopic eczema of varying severity.254–263 Five of these studies did not state whether an emollient was also used;254,258,259,262,263 two studies did not permit the use of emollients;256,260 in the remaining three studies emollients could be used as required.255,257,261 The comparisons were:

No trials compared moderately potent with mild potency topical corticosteroids. Few studies reported the quantities of topical corticosteroids used – where this information was given, the findings are summarised in this section.

Alclometasone dipropionate 0.05% (moderately potent) versus clobetasone butyrate 0.05% (moderately potent)

One double-blind RCT compared the effectiveness of alclometasone dipropionate 0.05% with clobetasone butyrate 0.05% (n = 43). In this small study, improvement in severity of signs and symptoms was not significantly different between groups. Investigator’s rating of the global condition was similar in both groups. Stinging was reported in two children treated with alclometasone.263 [EL = 1+]

Triamcinolone acetonide cream 0.1% (potent) versus hydrocortisone valerate cream 0.2% (moderately potent)

A within-patient (left–right) RCT compared 2 weeks’ triamcinolone acetonide 0.1% treatment with hydrocortisone 0.2% (n = 66, 54 completed and analysed). Severity was reported to be improved in both groups, but data were only shown in graphs. Clearance or an ‘excellent response’ was seen in 74% in both groups. Transient stinging was reported in 3% in both groups.283 [EL = 1−]

Hydrocortisone butyrate 0.1% cream (potent) versus hydrocortisone 1% ointment (mild)

One RCT evaluated two hydrocortisone preparations in a left–right comparison (hydrocortisone butyrate 0.1% cream versus hydrocortisone 1% ointment, n = 40). Treatment was given for 4 weeks. Significantly greater improvements in the global severity of the condition were reported in children treated with hydrocortisone butyrate 0.1%: mean (%) reduction in global severity score after 4 weeks of 2 (73%) versus 1.6 (62%), P < 0.05. Details of any adverse effects were not reported.258 [EL = 1+]

Betamethasone valerate 0.1% (potent) versus hydrocortisone 1% (mild)

One double-blind RCT compared the effectiveness of 3 days’ treatment with betamethasone valerate 0.1% with 7 days’ treatment with hydrocortisone 1% ointment in children with mild to moderate atopic eczema (n = 207).254 Treatment was used when needed during an 18 week period. The population consisted predominantly of children from the community in whom atopic eczema was milder than in the 16% recruited from a hospital outpatient clinic. Several outcomes were only reported for the community subgroup. After 18 weeks’ follow-up, no significant differences were found between groups in any outcome (scratch-free days mean difference 0.5 days, 95% CI −0.2 to 4.0 days, changes in quality of life scores (CLQI and DFI) or in the number of relapses or disturbed nights). Overall, 9% reported adverse events, which were mainly worsening of symptoms in 5% and 9% of the groups treated with the potent and mild topical corticosteroids, respectively. Other adverse events reported were cases of spots, rashes, hair growth and viral encephalitis in the group treated with betamethasone valerate 0.1%. Fifty percent of children who responded to a follow-up questionnaire expressed a preference for using a mild topical corticosteroid if it controlled the atopic eczema successfully. The remaining 50% of children who responded to the questionnaire expressed a preference for using a short burst of betamethasone because it reduced treatment time and controlled the atopic eczema quickly.254 [EL = 1+]

Mometasone furoate 0.1% (potent) versus various hydrocortisone preparations

Two RCTs compared mometasone furoate 0.1% with various hydrocortisone preparations in children with moderate to severe atopic eczema.

In the first study the comparator was hydrocortisone valerate 0.2% cream (moderate potency n = 219).256 The children had failed to respond to treatment with a hydrocortisone preparation (assumed to be a mild preparation) over the previous 7 days. It was reported that there were no significant differences between mometasone furoate 0.1% and hydrocortisone valerate 0.2% groups in global improvement (87% versus 78%, P = 0.01) after 3 weeks’ treatment. However, no baseline data were reported and thus it was not possible to determine whether groups were similar other than in the intervention being given.256 [EL = 1−]

In the second RCT the comparator was hydrocortisone 1% cream (n = 48). After 6 weeks’ treatment, significantly greater improvement in disease severity was reported in the mometasone group (95% versus 75% with hydrocortisone 1%, P = 0.01), and greater reduction in the total body surface area involved (reductions of 40% and 26%, respectively, P = 0.03). Overall, 63% in both groups discontinued treatment early owing to clearance of their condition. There was no significant difference between the two groups in mean morning plasma cortisol levels or in any changes in these levels, although numerical data were not reported.260 [EL = 1+]

Fluticasone propionate 0.05% (potent) versus hydrocortisone 1% (mild) or hydrocortisone 17-butyrate 0.1% (potent)

One publication reported the outcomes of two RCTs which compared fluticasone propionate 0.05% cream with hydrocortisone 1% (n = 137) or hydrocortisone 17-butyrate 0.1% (n = 128) in children experiencing a flare of atopic eczema.255 Treatment was applied twice a day for 2–4 weeks until the atopic eczema was stabilised, followed by intermittent use as required up to twice a day, for up to 12 weeks. Emollients could be used as required.

Greater improvement in total eczema score (a measure of three signs and the surface area affected) was reported with fluticasone compared with the hydrocortisone preparations in both studies at the end of both the acute and maintenance treatment phases. Also, significantly greater improvements in rash, itch and sleep disturbance were reported with fluticasone versus hydrocortisone 1%, and itch and sleep disturbance only with fluticasone versus hydrocortisone 17-butyrate 0.1%. Physicians considered that 84–98% of children had improved from baseline, the difference between groups being statistically significant for the fluticasone versus hydrocortisone 17-butyrate study. Time to recurrence was also reported, but no statistical analysis was presented. The quantities of topical corticosteroids used were similar in both studies. Adverse effects considered to be related to treatment were cases of folliculitis and tinea (ringworm), and development of red papules/boils with fluticasone; a case of flare with secondary infection with hydrocortisone 1%; and cases of itchy skin, minor skin infections/pustules, and impetigo on the face with hydrocortisone 17-butyrate 0.1%.255 [EL = 1+]

Triamcinolone acetonide 0.1% cream (potent) versus alclometasone dipropionate cream 0.05% (moderately potent)

One RCT compared triamcinolone acetonide 0.1% cream with alclometasone dipropionate 0.05% cream (n = 40). Treatment was used for up to 3 weeks. Improvements in severity of four signs and symptoms (erythema, lichenification, pruritus and exudation) were reported to be significantly greater with triamcinolone. Early-morning serum cortisol levels were measured in 68% of the children: no significant changes were reported, but mean differences from baseline were not quoted for the two treatment groups.257 [EL = 1+]

Mometasone furoate 0.1% (potent) versus clobetasone (ester not specified) 0.05% (moderately potent)

One RCT compared the effectiveness of mometasone furoate 0.1% and clobetasone 0.05% (n = 60). Mometasone was applied once daily and clobetasone twice daily. After 3 weeks’ treatment, there was significantly greater reduction in disease severity score with mometasone (86% versus 66% improvements, P < 0.01). The proportions of children with total clearance or improvement of the target area were: clearance 50% versus 7%, marked improvement 30% versus 37%, and moderate improvement 20% versus 50%. No adverse effects were reported during the trial.259 [EL = 1+]

Fluticasone propionate cream 0.05% (potent) versus clobetasone butyrate cream 0.05% (moderately potent)

One double-blind RCT compared fluticasone propionate 0.05% cream applied once daily with clobetasone butyrate 0.05% cream applied twice daily (n = 22).261 Treatment was given for up to 4 weeks, with an additional 2 weeks’ follow-up. There were no significant differences between groups in any outcomes (changes in SCORAD severity scores and 24 hour urinary cortisol excretion). In one child treated with clobetasone butyrate 0.05% cream, urinary cortisol excretion decreased during the study, but it had recovered by the follow-up visit.261 [EL = 1+]

Hydrocortisone butyrate 0.1% (potent) versus alclometasone dipropionate 0.05% (moderately potent)

One double-blind RCT compared the effectiveness of alclometasone dipropionate 0.05% to hydrocortisone 17-butyrate 0.1% (n = 40). Improvement in severity of signs and symptoms was not significantly different between groups after 2 weeks’ treatment. Investigator’s rating of the global condition was similar in both groups. Stinging was reported in two children treated with alclometasone and in one treated with hydrocortisone.262 [EL = 1+]

Comparisons with desonide (mild)

Two RCTs compared hydrocortisone 2.5% ointment or mometasone furoate 0.1% with desonide (a mild topical corticosteroid not available in the UK).284,285 These studies are considered in this section because they provided some safety data for hydrocortisone and mometasone.

After a mean of 27 days’ (maximum 42 days’) treatment with mometasone, ‘evidence of atrophy’ was reported in four children (17%); this was assessed by measuring the following signs on a four-point scale: thinning of the skin, striae, shiny skin, telangiectasia, loss of elasticity, and loss of normal lines on the cutaneous surface. Other adverse effects reported were burning on application in three children and appearance of fine hair in one child (n = 13).284 [EL = 3]

After 4 weeks’ treatment with hydrocortisone 2.5% ointment there were no significant differences in early-morning serum cortisol levels in response to an adrenocorticotrophic hormone (ACTH) test compared with baseline (mean change 1.3%) (n = 10).285 [EL = 3]

Different formulations of a topical corticosteroid of the same potency

Two within-person (left–right side) RCTs evaluated the global effectiveness and cosmetic acceptability of two different formulations of hydrocortisone 1% (an oil-in-water emulsion, and an ointment) in children with atopic eczema (total n = 156).268,269 Treatment was given for 4 weeks. Neither study reported baseline or demographic data, other than severity scores, and one did not report statistical analysis.269 The other found no significant difference between the two preparations in global improvement, but there was a significant difference in patient preference, with more preferring the emulsion than ointment.268,269 [EL = 1−]

Different frequency of application

The NICE TA programme issued guidance on the frequency of application of topical corticosteroids in 2004.12 The guidance applies to both children and adults with atopic eczema. The HTA informing the NICE guidance12 included three studies involving children, only two of which have been published in full.286,287 Data for the third study are reported in the HTA.288 No further RCTs considering frequency of application were identified.

The available studies compared once-daily with twice-daily application of clobetasone 17-butyrate 0.05% lotion (n = 30),287 fluticasone propionate 0.05% cream (n = 126)286 and fluticasone propionate 0.005% ointment (n = 120).288 The two trials involving fluticasone included both children and adults but data for children were reported separately. No significant differences were reported in outcomes following once- or twice-daily application of clobetasone 17-butyrate 0.05% lotion for 1 week, or fluticasone propionate 0.05% cream for 4 weeks. The RCT evaluating fluticasone propionate 0.005% ointment, which was reported only within the HTA, found that both investigator- and patient-rated success rates after 4 weeks’ treatment were significantly higher in the group using twice-daily application of the ointment.288 [EL = 1++]

Other studies of topical corticosteroids that focused on adverse effects

A post-marketing safety review of topical corticosteroids in paediatric patients (mean age 7.7 years) documented the adverse effects reported between 1987 and 1997 (n = 202).281 The body areas to which the topical corticosteroid was applied were the face and neck (20%), buttock, groin or genitals (16%), legs or feet (11%), arms or hands (10%), head or scalp (6%), trunk (4%), whole body (2%), or axillae (1%). The adverse effects occurring in 1% or more children were local irritation (33%), skin depigmentation or discolouration (15%), striae or skin atrophy (15%), Cushing syndrome (3%), growth retardation, hyperglycaemia, scarring and staphylococcal infection (each 2.5%), genital hypertrichosis, hirsutism and rosacea (each 2%), acne, glaucoma and hypersensitivity reaction (each 1.5%), and adrenal insufficiency, bruising, fungal infection, gynaecomastia, perioral dermatitis and mood change/’mental status’ (each 1%).281 [EL = 3]

Several case series or before-and-after studies considered the impact of topical corticosteroid treatment on adrenal function by measuring serum cortisol and/or ACTH levels.274–279,289 [EL = 3]

Two studies reported no significant changes in cortisol or ACTH levels or response to a short tetracosactride test after 1–4 weeks’ use of clobetasone butyrate 0.05% (total n = 41).274,275 [EL = 3]

No significant differences were found between pre- and post-treatment serum cortisol values (adrenal response to stimulation with tetracosactride) in children treated with fluticasone propionate 0.05% cream twice daily for up to 4 weeks (n = 51).276 Two children did not attain the usual response (minimum cortisol level) expected, and were considered to have adrenal suppression. Drug-related adverse effects reported were burning, urticaria, erythematous rash and telangiectasia.276 [EL = 3]

A safety study of fluticasone propionate 0.05% lotion (n = 44, age 3 months to 6 years) found no difference in cortisol levels after up to 4 weeks’ treatment compared with baseline in children with moderate to severe atopic eczema.290 [EL = 3]

One study compared serum cortisol levels in children treated with one of six different topical corticosteroids of different potencies (some not available in the UK): betamethasone dipropionate, diflucortolone valerate, halcinonide, clobetasone butyrate, desonide and flucortine butylester (n = 26).277 After 6 days’ treatment, plasma cortisol values decreased most from baseline with diflucortolone valerate (72%), followed by betamethasone dipropionate (61%), halcinonide (38%) and clobetasone butyrate (21%). Mean plasma cortisol values increased slightly with desonide and with flucortine butylester (1% and 15%, respectively). For those treated with diflucortolone, betamethasone and halcinonide, the cortisol levels fell below the normal range in 4/4, 4/5, and 2/4 children, respectively, during the first 6 days of treatment, and these levels normalised in 3/4, 2/4 and 2/2 children during continued treatment (no further details were provided). Of those treated with clobetasone, desonide or flucortine, none of the serum cortisol values fell outside the normal limits. These data should not be regarded as comparisons of the effects of the six products on cortisol levels, because as well as differences in potencies, the age of the children and the body surface area treated would influence systemic absorption of the topical corticosteroid, and these confounders were not accounted for in this study.277 [EL = 3]

Two cross-sectional studies compared adrenal response to a low-dose ACTH stimulation test in children with atopic eczema with the response in a control group. The children in both studies had been treated with topical corticosteroids since infancy. The first study included only children who had been treated with hydrocortisone 1% ointment (median duration 6.5 years, range 3–10 years; n = 28). None of the plasma cortisol measurements differed significantly between the two groups (basal, peak, increment or area-under-curve measurements).278 [EL = 3] The second study included children treated with topical corticosteroids of different potencies (median duration 6.9 years, range 0.5–17.7 years; n = 35).279 This study also reported no significant differences in adrenal response to ACTH between children treated with mild or moderately potent topical corticosteroids and controls. All four children treated with potent or very potent topical corticosteroids failed the ACTH test (failure was not defined; it was assumed that the ‘normal’ response was not attained).279

A retrospective study (n = 1271, 666 children) evaluated adverse effects to topical corticosteroids of various potencies, although it was not clear which products fell into which classification of potency used in the study.280 Treatment was used for at least 6 months. The cumulative incidence of several adverse effects increased with age (infants versus children): hypertrichosis (0.5% versus 1%), telangiectasia on cheeks (0% versus 2.3%), skin atrophy of antecubital or popliteal fossae (1.5% versus 5.2% and 1.9% versus 4.1%, respectively), acne and folliculitis (0% versus 1.3%), bacterial infection (1.4% versus 2.1%), and steroid-induced and contact dermatitis (0% versus 0.4% for both outcomes). There were no reports of striae atrophica. Cumulative incidence of fungal infection fell (1.9% versus 0.6%). The risk of telangiectasia on the cheeks appeared to be higher in those with longer duration of disease, and in those who applied more than 20 g to the face during the 6 month treatment period. The risk of atrophy of the antecubital and popliteal fossae was higher with longer duration of disease, and in those who used more than 500 g of topical corticosteroid during the treatment period.280 [EL = 3]

Topical corticosteroid versus a coal tar preparation

One within-patient (left–right side) RCT compared the effectiveness of a coal tar 1% cream to hydrocortisone 1% cream in children with dry, bilateral, symmetrical atopic eczema (n = 30).264 Treatment was used for 4 weeks. Use of emollients was not permitted. Confidence intervals for mean differences between groups in improvement in severity scores were not reported, but it was stated that there were no significant differences. This small study may have been underpowered to detect differences. No baseline data were reported (other than for severity scores), so it could not be determined whether groups were similar at baseline.264 [EL = 1−]

Topical corticosteroids versus topical calcineurin inhibitors

Evidence for this comparison is considered in Section 7.3.

Studies that have investigated the effectiveness of topical corticosteroids for preventing recurrence of flares are considered in Section 7.7.2. Studies that have investigated the effectiveness of topical corticosteroids in combination with antibiotics for treating infected atopic eczema are considered in Section 7.6.

Cost-effectiveness

No published economic evaluations of topical corticosteroids were identified. The NICE TA included an economic analysis on frequency of use of topical corticosteroids but the analysis did not distinguish between children and adults.288 The clinical outcomes were reported in the TA to be equivalent and thus the cost-effectiveness analysis was an analysis of costs of treatment only.

The TA stated that where there is no clear difference in clinical outcome by frequency, the choice of treatment should be guided by cost per patient treated, taking into account product costs at that point in time and frequency of use. The TA concluded that, given the small cost difference between regimens, any treatment would be highly likely to be cost-effective if it could demonstrate better outcomes than other topical corticosteroid treatments. Also, better outcomes would be likely to reduce the need for additional GP visits to address problems associated with treatment failure.

The cost savings associated with once-daily treatment were calculated using various scenarios (number of flares per year and quantities of topical corticosteroid used and wasted). However, given the lack of clinical evidence for this, or any other basis on which to make a reasonable judgement on the percentage of products used and wasted in any treatment period, the TA was not able to conclude with any certainty whether once-daily use of topical corticosteroids would lead to cost savings for the NHS.

Since no economic evaluation studies were identified that considered the cost-effectiveness of topical corticosteroids of different potencies, it was not possible to assess whether the additional number of successful treatments using topical corticosteroids of higher potency were ‘worth’ the additional costs associated with treatment, taking into account the small risk of harmful side effects associated with more potent topical corticosteroids.

Evidence statement for topical corticosteroids

Few trials have evaluated topical corticosteroids in a way that reflects their use in UK practice (that is, management of flares/exacerbations in children already using emollients). RCTs that compared 2–4 weeks’ treatment with a topical corticosteroid with vehicle in children and adults generally reported a greater response rate in the topical corticosteroid group, although a noticeable effect of vehicle (emollient) was apparent. [EL = 1−] Greater efficacy was seen in an RCT comparing an emollient used with a mild topical corticosteroid with the topical corticosteroid used alone (one trial). [EL = 1+]

In comparisons of two formulations of mild topical corticosteroids, there were differences in patient preference but no differences in clinical outcomes. [EL = 1−] No significant differences were identified between two moderately potent preparations (one trial). [EL = 1+]

Compared with mild preparations, potent topical corticosteroids generally led to significantly greater improvements in outcomes (severity and global improvements) following 2–6 weeks’ treatment, although only one of the available studies evaluated quality of life. [EL = 1+] The outcome of 3 days’ treatment with betamethasone valerate 0.1% (potent) was not significantly different to 7 days’ treatment with hydrocortisone 1% (mild) in one trial involving children with mild to moderate atopic eczema treated mainly in the community. [EL = 1+] No consistent differences in effectiveness between moderately potent and potent topical corticosteroids were evident from the available data. A comparison of two potent preparations found some differences between the preparations in some outcomes (one trial). [EL = 1+] No evidence of the cost-effectiveness of different potencies of topical corticosteroids was identified.

Once-daily and twice-daily application of topical corticosteroids are both effective for the treatment of atopic eczema. It is not possible to distinguish between them on effectiveness or cost-effectiveness grounds. [EL = 1++]

Several studies reported changes in serum cortisol levels or responses to adrenal stimulation following topical corticosteroid treatment. It appeared that short-term use of topical corticosteroids of any potency did not cause statistically significant or clinically important suppression of adrenal function. In children treated with mild topical corticosteroids for several years, no evidence of adrenal suppression was found compared with a control group (one study). While there was some suggestion that adrenal suppression could occur with potent topical corticosteroids, the available studies were not designed nor sufficiently powered to address what quantities or duration of use affected the risk of adrenal suppression. [EL = 3]

Other adverse effects reported with topical corticosteroids across the available studies included stinging on application, hypertrichosis, telangiectasia on cheeks, skin atrophy of antecubital or popliteal fossae, acne, folliculitis, bacterial infection, and steroid-induced and contact dermatitis. [EL = 3]

The single trial that compared a coal tar preparation with hydrocortisone 1% was of poor quality and did not allow any conclusions to be drawn. [EL = 1−]

From evidence to recommendations

The order in which emollients and topical corticosteroids should be applied is not known. Mixing creams and ointments may change the properties (formulation and absorption characteristics) of the treatments. The GDG therefore believes that a short interval (several minutes) should be left between application of a topical corticosteroid and an emollient, where practicable. [EL = 4] The GDG believes that treatments should be applied at times of day that are convenient to the child and their parents or carers. This could mean applying one dose of topical corticosteroid before school and one after school so that emollients are the only treatments needed at school.

It is the GDG’s view that a short treatment with a potent topical corticosteroid is as effective as a longer treatment with a mild preparation. [EL = 4]

In children with frequent recurrent flares (two or three per month) of atopic eczema, the GDG believes that topical corticosteroids can be used for two consecutive days per week as a strategy for flare prevention. This is sometimes referred to as weekend therapy. This strategy can only be started once a flare has been controlled.

The risk of adverse effects due to topical corticosteroids is related to the surface area to which they are applied, the thickness of the skin, potency and duration of use. Therefore it is the GDG’s view that treatment should be applied only to affected areas unless weekend therapy is being used in chronic persistent areas to prevent flares. The face and neck should only be treated with mild topical corticosteroids, apart from severe flares where topical corticosteroids of moderate potency can be used for up to 5 days. Moderately potent or potent topical corticosteroids can be used on other areas of thin skin such as the axillae and groin for 7–14 days only. [EL = 4]

The GDG believes that many healthcare professionals do not have enough specialist dermatological knowledge to recognise minimal signs of infection. Infection can be a cause of worsening or uncontrolled atopic eczema and it is important for healthcare professionals to consider infection before stepping up treatment to potent topical corticosteroids. [EL = 4]

Withholding topical corticosteroid treatment may lead to worsening of the child’s atopic eczema, and deterioration in the child’s quality of life. Adverse effects rarely occur when topical corticosteroids are used appropriately.

The GDG believes that topical corticosteroid preparations should be labelled with their potency group, and that this label should be applied to the container rather than the outer packaging to avoid confusion over potency, in order that the directions for use are not lost.

Recommendations for topical corticosteroids (including research recommendations) are presented in Section 7.11.

7.3. Topical calcineurin inhibitors

Pimecrolimus and tacrolimus are topical immunosuppressants. Pimecrolimus is derived from a fungus called Streptomyces hygroscopicus and tacrolimus is derived from Streptomyces tsukubaensis. Both pimecrolimus and tacrolimus bind to and inhibit the action of a protein called calcineurin, which is involved in the activation of T cells (one of the cell types that become activated in the skin of people with atopic eczema). They are therefore called calcineurin inhibitors. The main effect of calcineurin inhibitors is to inhibit the production of cytokines (chemical messengers) produced by the T cells, which lead to the inflammation that produces flares of atopic eczema.

Topical tacrolimus ointment is available in two strengths: 0.03% and 0.1%. Only the 0.03% ointment is licensed for use in children, and this may only be prescribed for children aged 2 years or over. Pimecrolimus is a 1% cream that is licensed for use in children aged 2 years or over.

Overview of available evidence

NICE guidance on topical tacrolimus and pimecrolimus for the treatment of atopic eczema in children and adults was published in 2004.13 The HTA that informed the NICE guidance included evidence for both children and adults.291 Evidence that relates to children is summarised in this section, together with evidence published more recently. The HTA included the following RCTs in children:

The following additional studies have been published since the HTA:

Except for one RCT, all were funded by the manufacturers of the calcineurin inhibitors, and they tended to be of similar design, evaluating the same outcomes. Note that several studies were reported in more than one publication.

Pimecrolimus

Studies included in the HTA

The studies included in the HTA were vehicle-controlled double-blind RCTs evaluating pimecrolimus 1% cream in children aged 1–17 years, the majority aged 12 years or under. The first study report pooled outcome data for children with mild or moderate atopic eczema who were also treated with emollients from two identical RCTs (n = 403).295 After 6 weeks’ treatment there were significant differences in efficacy outcomes between the pimecrolimus 1% and vehicle groups:

  • 35% versus 18% (pimecrolimus 1% cream versus vehicle) were clear or almost clear (IGA score of 0 or 1) of atopic eczema, P ≤ 0.05
  • severity scores (EASI) fell by 45% versus 1%, P ≤ 0.001
  • 55% versus 33% had a pruritus score of none or mild itching/scratching, P < 0.001
  • 61% versus 40% of parents reported good or complete control of the child’s atopic eczema, P < 0.05.

The effects of treatment on the quality of life of parents of children aged up to 8 years in this study were reported in a separate publication (n = 278). Data from 80% at 6 weeks showed significantly greater improvements in PIQoL-AD scores in the pimecrolimus 1% group compared with vehicle (least squares mean change −3.2 versus −1.63, difference 1.57, 95% CI 0.22 to 2.92).296 No significant differences were found in any of the reported adverse effects. Overall, 10.4% in the pimecrolimus group and 12.5% in the vehicle group had application-site reactions.295 [EL = 1+]

The second RCT considered the effectiveness of pimecrolimus 1% cream in the prevention of flares in children with mild to moderate atopic eczema (n = 713).297 Treatment with pimecrolimus or vehicle was applied at the first sign (erythema) or symptom (pruritus), to prevent progression to flare. A flare was defined as at least severe erythema and severe infiltration/papulation (IGA score of 4 or more). Emollients were used throughout the study by both groups, and both groups also applied a moderately potent topical corticosteroid during flares. Significantly fewer children experienced flares in the pimecrolimus 1% group at both 6 months (39% pimecrolimus versus 66% vehicle, P < 0.001) and 12 months (49% versus 72%, P < 0.001); relative risk (RR) of having a flare with pimecrolimus 1% compared with vehicle at 12 months 0.69 (95% CI 0.61 to 0.77). Fewer children treated with pimecrolimus used topical corticosteroids for flares than those receiving vehicle (43% versus 68%), and the mean proportion of days spent being treated with topical corticosteroids was 4% versus 9%. Of the adverse effects reported, no significant differences were seen between groups except in the incidence of viral infection (12.4% pimecrolimus versus 6.3% vehicle). More children withdrew from the vehicle arm (51.5% versus 31.6%), which was predominantly due to an unsatisfactory therapeutic response.297 [EL = 1+]

Studies published since the HTA

The use of pimecrolimus 1% cream was evaluated in children aged 3–23 months in two vehicle-controlled double-blind RCTs of 4–6 weeks’ duration.110,299–301 Treatment was applied twice daily to affected areas. Emollients were permitted on unaffected areas throughout both trials.

One study found that, at 6 weeks, the proportions of children with IGA scores of clear or almost clear were significantly higher in the pimecrolimus 1% group (55% versus 24% with vehicle, P < 0.001). Improvements in severity (EASI score), the proportions of children with absent or mild pruritus, or with a carers’ assessment of complete or good control were also significantly greater with pimecrolimus. Other than a significant difference in the incidence of pyrexia (32% pimecrolimus 1% cream versus 13% vehicle), there were no other differences in adverse effects between groups. The discontinuation rates in the pimecrolimus 1% and vehicle groups were 11% and 48%, respectively. [EL = 1+] Following the 6 week double-blind period, all children were offered treatment with pimecrolimus. Overall, 93% used pimecrolimus 1% cream for a further 20 weeks. The data suggested sustained benefit. All adverse effects reported in both groups were common childhood ailments (including pyrexia, nasopharyngitis and otitis media). Pyrexia was the only adverse effect that occurred in significantly different proportions in treatment groups (32% pimecrolimus 1% cream versus 13% vehicle, P < 0.05).299 [EL = 3]

The second study reported significantly greater improvements in EASI, IGA and SCORAD scores in children treated with pimecrolimus 1% cream for 4 weeks compared with placebo (n = 196). There were no significant differences between groups in the change in the proportion of children with dry skin, or in adverse effects.300,301 [EL = 1+] Quality of life outcomes at 4 weeks were reported in a separate publication (quality of life in parents of children with atopic dermatitis (PQoL-AD)). Significantly greater improvements in each of the five subscales were reported in those treated with pimecrolimus compared with vehicle (psychosomatic wellbeing, effects on social life, confidence in medical treatment, emotional coping, acceptance of disease).110 Following the randomised phase of the study, children were offered pimecrolimus treatment for 12 weeks. During this time, improvements in efficacy outcomes were reported to be sustained although no numerical data were reported. Adverse effects believed to be related to treatment (which treatment was not specified) occurred in six children (two cases of impetigo, and one case each of eczema herpeticum, varicella, asthma, aggravated atopic eczema, and exacerbated eczema).301 [EL = 3]

Two RCTs considered the effectiveness of pimecrolimus 1% cream compared with vehicle in the prevention of flares.302–304 Emollients were used in both studies to treat dry skin. The first included children aged 3–23 months (n = 250).302 The study was identical in design to one in older children described earlier.297 Significantly fewer children experienced flares in the pimecrolimus group at 6 months (32% pimecrolimus 1% cream versus 70% vehicle) and at 12 months (43% versus 72%); the mean numbers of flares per child were 1.0 versus 2.2, P < 0.001. Fewer children treated with pimecrolimus 1% used topical corticosteroids for flares than those receiving vehicle (36% versus 63%), and the mean proportion of days spent being treated with topical cortico-steroids was 3% in the pimecrolimus group and 6% with vehicle (which corresponds to 11 days’ use and 22 days’ use, respectively). There were no significant differences between groups in the proportion with an IGA score of clear or almost clear, in severity (EASI) or pruritus scores or caregivers’ assessment at 12 months. There were no significant differences in the incidence of the reported adverse effects (application-site reactions or skin infection).302 [EL = 1+] Overall, 91 children (36%) continued into a second year of the study, applying pimecrolimus 1% for a median of 99 days. The data indicated sustained response to pimecrolimus 1% and no increase in incidence of adverse effects.303

A further RCT considered the effectiveness of pimecrolimus 1% cream in preventing progression of atopic eczema to flares in children aged 3 months to 11 years (n = 275).304 Pimecrolimus 1% or vehicle was used at the first signs or symptoms of atopic eczema. If after 7 days’ treatment with pimecrolimus or vehicle the child was believed to have a major flare, the evening dose of pimecrolimus 1% or vehicle was substituted with a potent topical corticosteroid. After 6 months’ treatment, significantly more children in the pimecrolimus 1% group had not experienced a flare (52% versus 34% with vehicle, P = 0.007). Time to first flare and the median time between first and second flares were also significantly longer in the pimecrolimus 1% group. Mean duration of use of topical corticosteroids was 10.9 days with pimecrolimus 1% and 17.3 days with vehicle, P = 0.002. The withdrawal rate due to unsatisfactory therapeutic effect was significantly higher in the vehicle group (14.3% versus 3.8%, P = 0.003). Rhinorrhoea (runny nose) was the only adverse effect reported in significantly different proportions between groups (9.8% pimecrolimus 1% cream versus 2.2% vehicle, P = 0.025). Other reported adverse effects were predominantly respiratory or gastrointestinal.304 [EL = 1+]

Quality of life data from two RCTs297,302 that considered whether pimecrolimus 1% cream prevented flares have been published separately in a single report.109 Both studies considered quality of life of parents of children aged up to 8 years (using PIQoL-AD), and one considered the quality of life of children aged 5 years or over (using CDLQI). Improvements in both measures were significantly greater with pimecrolimus 1% compared with vehicle.109

Pooled analysis of pimecrolimus versus vehicle studies

Data from three vehicle-controlled RCTs295,299 were pooled in one report in order to consider the treatment effects in children of different ethnicities (n = 589). Children were subdivided into those of Caucasian origin (54%) and non-Caucasian origin (46%, of which 42% were black, 12% Asian, and 47% ‘other’, mainly Hispanic). No significant differences in treatment response (IGA and EASI scores) or in application-site reactions were found between children of Caucasian or non-Caucasian origin.306 [EL = 1+]

Case series of pimecrolimus

A case series reporting the use of pimecrolimus 1% cream in children and adults with atopic eczema included some data for children aged under 2 years and those aged 2–12 years (n = 591 (62%) aged 2–12 years). Pimecrolimus 1% was applied to affected areas twice daily at the first signs or symptoms of atopic eczema. Other ‘usual treatments’ were permitted at the physician’s discretion. Of all patients enrolled, 88% used emollients at baseline; 53% used a topical corticosteroid at least once during the study; and pimecrolimus was used for 75% of the time, and daily by 55%. In children, improvements in IGA whole-body and facial scores were reported in 66% and 78%, respectively, for those aged under 2 years, and in 71% and 79%, respectively, of children aged 2–12 years. The most common adverse effects (reported in more than 10% of children aged up to 12 years) were nasopharyngitis, upper respiratory tract infection, cough and pyrexia. Overall, 5.2% reported application-site burning, and 2% reported worsening of atopic eczema. Treatment-related adverse effects reported in children were five cases (0.8%) of eczema herpeticum.315 [EL = 3]

Three case series measured blood concentrations of pimecrolimus following application of the 1% cream. The first found that, of 100 samples taken after 10 days’ treatment, the blood concentration of pimecrolimus was below 2 ng/ml in 96%, and the difference in mean concentration between those with 90% and those with 10% of body surface area affected was 0.4 ng/ml (n = 22).316 In the second study, the concentration of pimecrolimus was below 2 ng/ml in 98% of samples taken on days 4 and 22 of treatment. Results were in a similar range on days 4 and 22. The mean difference in blood concentrations between those with 90% and those with 10% of body surface area being treated was 0.7 ng/ml; on linear regression analysis a significant increase in blood concentrations with increasing surface area was found, P = 0.28 (n = 26).317 Five infants (6–12 months of age) from the latter study were followed up for 1 year, with a mean duration of use of pimecrolimus of 332 days. Mean blood concentrations were 0.32 ng/ml at week 27, and 0.68 ng/ml at week 53.318 [EL = 3]

Tacrolimus

Studies included in the HTA

Four RCTs included in the HTA evaluated the use of tacrolimus 0.03% ointment in children.265,266,292–294 Three of these also compared tacrolimus 0.03% ointment with higher strengths (0.1% and/or 0.3%) of topical tacrolimus.

One RCT compared 3 weeks’ treatment with three strengths of topical tacrolimus ointment with vehicle in children aged 7–16 years (n = 180). Children were also permitted to use emollients on unaffected areas. All strengths of tacrolimus ointment (0.03%, 0.1% and 0.3%) led to significantly greater improvements in effectiveness compared with vehicle (physician’s and patient’s global evaluations, EASI, head and neck score, and pruritus). No significant differences in incidence of application-site reactions (burning, pruritus or erythema) were reported. Blood concentrations of tacrolimus appeared to increase with increasing strength of the ointment applied.292 [EL = 1+]

Another RCT compared topical tacrolimus 0.03% and 0.1% ointment with vehicle in children with moderate to severe atopic eczema (n = 351).293,294 Treatment was applied twice daily for up to 12 weeks, or less if the atopic eczema cleared sooner. Emollients were permitted on unaffected areas. Both strengths of tacrolimus ointment were significantly more effective than vehicle in all effectiveness outcomes (physician’s and patient’s global assessment, changes in EASI and pruritus scores, body surface area affected, and quality of life (CDLQI)). The incidence of skin burning, pruritus, varicella, and vesiculobullous rash was significantly higher with tacrolimus 0.03% ointment compared with vehicle.293,294 Blood tacrolimus concentrations were measured: none was detected in 90%, and mean and median levels were below the limit of quantification (2 ng/ml) at all time points.293 [EL = 1+]

One RCT compared the effectiveness of tacrolimus 0.03% ointment applied once or twice daily with hydrocortisone acetate 1% in children with moderate to severe atopic eczema (n = 624). Treatment was given for 3 weeks. Use of unmedicated emollients and bath oils was permitted. Tacrolimus 0.03% ointment (applied once or twice daily) was significantly more effective than hydrocortisone 1% in changes in severity scores (modified EASI (including assessment of itch) and EASI); twice-daily application of tacrolimus 0.03% ointment was also significantly more effective than once-daily application in this outcome. Analysis of between-group differences in physician’s or parent’s/child’s global assessment, itch or sleep quality was not reported. The incidence of skin burning was significantly higher in both tacrolimus 0.03% ointment groups compared with hydrocortisone acetate 1% (23.2% with once-daily application of tacrolimus, 23.8% with twice-daily application, and 14.5% with hydrocortisone, P = 0.028). No other significant differences were found in the most commonly reported adverse effects (pruritus, folliculitis, influenza syndrome, skin infection).265 [EL = 1+]

Another RCT compared the effectiveness of tacrolimus 0.03% and 1% to hydrocortisone acetate 1% in children with moderate to severe atopic eczema (n = 560). Treatment was applied twice daily for 3 weeks. Use of bath oils and unmedicated emollients was also permitted. Median improvements in modified EASI scores (including an assessment of itch) were significantly greater with both tacrolimus ointment groups compared with placebo, and with tacrolimus 0.1% versus 0.03% (55.2% tacrolimus 0.03%, 60.2% tacrolimus 0.1%, 36% hydrocortisone, P < 0.001 tacrolimus groups versus hydrocortisone and P = 0.006 tacrolimus 0.1% versus 0.03%). The proportion of children with a physician-rated improvement of 90% or more was significantly higher in both tacrolimus groups compared with hydrocortisone (38.5% tacrolimus 0.03%, 48.4% tacrolimus 0.1%, 15.7% hydrocortisone, P = 0.001 both tacrolimus ointment groups versus hydrocortisone and P = 0.055 between tacrolimus groups). Skin burning occurred in significantly more tacrolimus-treated children compared with hydrocortisone (18.5% tacrolimus 0.03%, 20.4% tacrolimus 0.1%, 7% hydrocortisone, P < 0.05 both tacrolimus groups versus hydrocortisone). No other significant differences in the incidence of adverse effects were reported (pruritus, folliculitis, skin infection and skin erythema).266 Blood concentrations of tacrolimus were measured. Overall, 1.3% of all measurements were 1 ng/ml or higher in those treated with tacrolimus 0.03%, compared with 11.3% in the group treated with tacrolimus 0.1%.266 [EL = 1+]

Studies published since the HTA

One RCT compared tacrolimus 0.03% ointment with vehicle (both applied twice daily) in children with mild to moderate atopic eczema (n = 317).307 Unmedicated emollients were permitted on unaffected areas. After 6 weeks’ treatment, improvements in all efficacy outcomes were significantly greater in the tacrolimus group (IGA, body surface area affected, EASI and itch scores). Itching and erythema occurred in significantly more children treated with vehicle than tacrolimus (itching 23.4% versus 33.3%, P = 0.05; erythema 7.6% versus 18.9%, P = 0.003), and the withdrawal rate due to skin reactions was also significantly higher in the vehicle group (2.5% versus 7.5%, P = 0.04). There were no other significant differences in adverse effects reported (burning/stinging, folliculitis, skin infections, acne and eczema herpeticum).307 [EL = 1+]

One RCT aimed at comparing application-site reactions between topical tacrolimus 0.03% ointment and pimecrolimus 1% cream in children with moderately severe atopic eczema.305 Emollients were permitted on unaffected areas. At day 4, the proportions of application-site reactions were 26% with tacrolimus 0.03% ointment and 24% with pimecrolimus 1% cream. Erythema/irritation occurred in 19% versus 8% (P = 0.039), itching in 20% versus 8% (P = 0.073) and warmth/stinging/burning in 17% versus 20% (P = 0.931). Withdrawal rates were 4% with tacrolimus and 18% with pimecrolimus. No significant differences were reported in efficacy outcomes assessed at 6 weeks (proportions of children with IGA scores of clear or almost clear 42% versus 30%, P = 0.119; proportions of children with absent or mild pruritus 70% versus 64%, P = 0.493).305 [EL = 1+]

One RCT compared tacrolimus 0.03% ointment with clobetasone butyrate 0.05% cream and with combined use of the two preparations in children aged 7–15 years with moderate to severe atopic eczema (n = 45).267 Treatment was applied twice daily except in the combination group where tacrolimus was applied in the morning and clobetasone butyrate in the evening. Use of unmedicated emollients and bath oils was permitted. After 4 weeks’ treatment, improvements in modified EASI scores and the reduction in body surface area affected were significantly greater with clobetasone butyrate than tacrolimus 0.03% ointment, and with combination therapy compared with tacrolimus ointment alone. No between-group analysis was reported for IGA. Differences in skin burning rates between groups were not statistically significant.267

One RCT compared 0.03% tacrolimus ointment with topical methylprednisolone in 265 children (mean age 7.5 ± 4.2 years tacrolimus, 7.8 ± 4.2 years methylprednisolone) with severe to very severe atopic eczema.308 [EL = 1−] Children were randomised to either tacrolimus 0.03% ointment applied twice daily or methylprednisolone 0.1% in the evening over all affected areas for 2–3 weeks. Cleared areas were treated for an additional 7 days post clearance. At the end of the study, IGA and EASI scores and body surface area affected all showed significant improvement in both groups, with no statistically significant differences between the groups. Children’s assessment of itch (P = 0.0004) and sleep (P = 0.0094) on a visual analogue scale were significantly better in the methylprednisolone group than in the tacrolimus group. The study also highlighted the difference in mean cost of the treatment used (tacrolimus 100.99 euros versus methylprednisolone 14.59 euros).

In a cohort study which used within-patient (left–right side of body) comparison, tacrolimus 0.03% or 0.1% ointment was compared with the child’s usual topical corticosteroid treatment.309 [EL = 2−] Ninety-six children (aged 12 years or under) with moderately severe atopic eczema were treated on one side of their body (arms and legs) with their usual topical corticosteroid and on the other side with tacrolimus 0.03% for 7 days. If the tacrolimus 0.03% had no effect in the first 7 days the dosage was increased to tacrolimus 0.1% for a further 7 days. After the first 7 days 48/93 children had a greater improvement with tacrolimus 0.03% compared with the topical corticosteroid as determined by clinical assessment of erythema and lichenification. The other 45 children were then given tacrolimus 0.1% for the same side of the body for another week. Over this second period of treatment 24/45 showed a more marked improvement compared with the usual treatment side. Thus, overall tacrolimus treatment (0.03% and/or 0.1%) showed greater improvement in 77% of the children treated compared with their usual topical corticosteroid.

Case series of tacrolimus

Four case series reported adverse effects in children and adults who had used topical tacrolimus over longer periods than evaluated in RCTs. The majority of patients used tacrolimus 0.1% ointment.310–313 [EL = 3]

Three case series reported the most common adverse effects (occurring in 5% or more) in children aged 2–15 years who had been treated with topical tacrolimus for 6 months (n = 236, 35% were children),311 34 weeks (n = 3959 children)310 and 16 months (n = 466 children).313 Application-site effects were burning (19–38%),310,311,313 pruritus (17–34%),310,311,313 skin infection (15–32%),310,311,313 paraesthesia (numbness 9%),311 warmth (5%)311 and skin erythema (5–7%) 310,313 [EL = 3]

The fourth case series provided data for children aged 2–15 years who were treated with tacrolimus 0.1% ointment for a median of 902 days (2.5 years; range 1–1186 days). The most common application-site events (occurring in 5% or more) were pruritus (21% in children aged 2–6 years and 19% in those aged 7–15 years), pustular rash (15.7% and 11.2%, respectively), skin burning (20.5% and 18.0%, respectively), skin erythema (10.8% and 5.8%, respectively) and skin infection (22.7% and 22.3%, respectively). The incidence of infections in children aged 2–5 years and those aged 7–15 years was herpes simplex 4.3% and 6.3%, respectively, warts 6.5% and 7.3%, respectively, varicella zoster 9.2% and 1.9%, respectively, molluscum contagiosum 3.2% and 4.9%, respectively, and eczema herpeticum 0% and 0.5%, respectively. Discontinuation rates due to adverse effects were 2.7% in children aged 2–6 years and 1.0% in children aged 7–15 years.312 [EL = 3]

A fifth case series investigating the effect of tacrolimus 0.03% on moderate to severe atopic eczema in children (n = 58, mean age 6.98 ± 2.81 years) over a 4 week period showed a statistically significant improvement from baseline in the severity of the atopic eczema (EASI) and quality of life (CDLQI) (P < 0.001 and P < 0.01, respectively).314 [EL = 3] Adverse events reported were similar to the other case series of longer duration and higher dose of tacrolimus, namely burning, erythema and itching.

Other relevant guidance

As well as NICE guidance, a Europe-wide safety review of the risks and benefits of topical tacrolimus and pimecrolimus ointments was completed in March 2006, following reports of malignancy (skin cancers, lymphomas and others) in association with the use of these two products.319 The conclusion was that a causal link could not be determined. The UK Medicines and Healthcare products Regulatory Agency (MHRA) advised that:

  • pimecrolimus 1% cream should be used as a second-line treatment for mild to moderate atopic eczema where treatment with topical corticosteroids is not possible or inadvisable
  • tacrolimus ointment remains as a second-line treatment for moderate or severe atopic eczema in patients who do not have an adequate response to, or are intolerant of, topical corticosteroids
  • treatment with pimecrolimus or tacrolimus should only be initiated by physicians experienced in the diagnosis and treatment of atopic eczema; they should not be given to patients with congenital or acquired immunodeficiencies, or to patients on therapy causing immuno-suppression; and they should not be applied to malignant or potentially malignant skin lesions
  • neither pimecrolimus 1% cream nor tacrolimus 0.03% ointment is licensed for use in children aged under 2 years
  • in children the frequency of administration of tacrolimus 0.03% ointment should be limited to once daily
  • the lower strength of tacrolimus should be used in adults wherever possible
  • the products should be applied thinly and to affected areas of skin only
  • treatment should be short term; continuous use should be avoided
  • if no improvement occurs (after 6 weeks’ pimecrolimus treatment or 2 weeks’ tacrolimus treatment), or if the disease worsens, the diagnosis of atopic eczema should be re-evaluated and other therapeutic options considered.319

Evidence statement for topical calcineurin inhibitors

In short-term studies (4–6 weeks), pimecrolimus was more effective than vehicle alone in children with mild to moderate atopic eczema in terms of physician-reported measures of disease activity (including global assessment of disease activity, reduction in severity and itching), and improvements in quality of life of children and their parents. [EL = 1+] Intermittent application of pimecrolimus at the first sign or symptom of atopic eczema was more effective than continuous application of emollients in reducing the frequency of flares and the need for concomitant use of topical corticosteroids to treat flares, and in improving quality of life of parents and children. [EL = 1+] While most adverse effects reported occurred with similar frequency with pimecrolimus and vehicle, the incidence of viral infections, pyrexia and rhinorrhoea (runny nose), all of which are common childhood ailments, was significantly higher with pimecrolimus (one study each). Skin infections believed to be associated with pimecrolimus use included varicella, herpes simplex and eczema herpeticum. Application-site reactions were common with both pimecrolimus and vehicle, and not significantly different in overall incidence between pimecrolimus and tacrolimus (one study). [EL = 1+] No studies that compared pimecrolimus with topical corticosteroids were identified.

In short-term studies (3–12 weeks), tacrolimus 0.03% ointment was more effective than vehicle alone in children with mild to severe atopic eczema in terms of physician-reported measures of disease activity (including global assessment of disease activity, reduction in severity and itching) and improvement in children’s quality of life. Twice-daily application of tacrolimus was more effective than once-daily application in reducing severity in children with moderate to severe atopic eczema (one study). [EL = 1+] Tacrolimus use was more commonly associated with skin burning, and greater skin erythema/irritation than pimecrolimus was (one study). [EL = 1+] Compared with a mild topical corticosteroid (hydrocortisone acetate 1%), tacrolimus 0.03% and 0.1% ointments were both more effective in reducing disease severity in children with moderate to severe atopic eczema. [EL = 1+] Differences between tacrolimus 0.03% and 0.1% were inconsistent. Evidence from one small trial suggested that short-term use of a moderately potent topical corticosteroid (clobetasone butyrate 0.05%) alone or in combination with tacrolimus 0.03% ointment was more effective than tacrolimus 0.03% ointment alone in reducing severity and body surface area affected by atopic eczema. [EL = 1+] There was a lack of data for tacrolimus compared with potent topical corticosteroids.

Cost-effectiveness

Studies included in the HTA

The HTA291 that informed the NICE TA13 reviewed the cost-effectiveness of tacrolimus and pimecrolimus for various severities of atopic eczema. Only one published cost-effectiveness analysis (which considered both costs and effectiveness simultaneously rather than costs alone) was identified in the HTA review.320 This US study compared the cost-effectiveness of a course of tacrolimus with 2 week and 4 week courses of topical corticosteroids. The study was poorly conducted (it failed to use appropriate methods for calculating cost-effectiveness ratios) and so the cost-effectiveness analysis was recalculated in the HTA using data from the published study. The results showed that tacrolimus was dominant, that is it was both less costly and more effective than the 2 week course of topical corticosteroids, but the 4 week course of topical corticosteroids was more cost-effective than either tacrolimus regimen. The reported costs were modest (US$7 for tacrolimus, US$10 for the 2 week course of topical corticosteroids and US$7 for the 4 week course), but this was of very limited relevance in the context of the NHS.

Two economic models from pharmaceutical industry submissions were also reviewed in the HTA. The tacrolimus model did not measure benefits in QALYs, and the pimecrolimus model compared treatment with placebo only and thus was of very limited value.

A model was developed for the HTA to evaluate the cost-effectiveness of pimecrolimus and tacrolimus for children and adults in the UK. The pimecrolimus analysis was also reported separately in a subsequent publication.321

Eight Markov (state transition) models representing specific cohorts of adults and children (aged 2–16 years) were created. Each group was modelled separately in order to calculate the costs and outcome values associated with that group. The four children’s models were for those with:

The treatment alternatives considered were:

Cost data were derived from data for the NHS published in 2003. Cost of infections and out-of-pocket expenses were not included since there was no evidence that these would differ across the two arms of the trials. Outcomes were expressed in QALYs, with QALY weightings derived from decrements in IGA scores (0.86 for an average of 0–1 decrements or ‘mild’ disease, 0.69 for 2–3 decrements or ‘moderate’ disease, and 0.59 for 3–5 decrements or ‘severe’ disease).

The results showed that there were fewer benefits (QALYs) associated with using pimecrolimus for mild to moderate body and facial atopic eczema in children as first- and second-line treatment relative to topical corticosteroids alone, and that there were higher costs associated with pimecrolimus. Therefore topical corticosteroids were the most cost-effective option for children with mild to moderate disease.

For the treatment of children with moderate to severe body eczema, tacrolimus conferred some additional health benefits over topical corticosteroids, but with an incremental cost per QALY of around £9,000 as first-line therapy and £14,000 as second-line therapy. This appeared to indicate that tacrolimus is cost-effective (below the NICE threshold for cost-effectiveness of £20,000 per QALY). However, the authors reported that these results were highly sensitive to changes in assumptions in the model, meaning that the results are not very robust.

For the treatment of moderate to severe facial eczema, the additional cost per QALY of tacrolimus as second-line therapy compared with topical corticosteroids was around £36,000 and as first-line therapy it was dominated by topical corticosteroids (that is, it was both more expensive and had fewer benefits). The results were highly sensitive to changes in the model parameters, making it difficult to draw conclusions about the relative cost-effectiveness of the treatment options.

Additional probabilistic analyses were undertaken in the HTA by simulating 1000 trials of the three treatment options to assess the likelihood of any of them being cost-effective. These analyses all indicated that the probability of any treatment option being cost-effective was less than 50%, reflecting the considerable uncertainty of the results.

The NICE TA interpreted this evidence taking into account additional analysis undertaken by the manufacturers of tacrolimus and pimecrolimus. It concluded that the cost-effectiveness analyses undertaken by the HTA indicated similar outcomes for each of the treatment strategies and that the uncertainty of specific variables used in the models meant that the results of the economic analysis could not form the basis of their recommendation. The manufacturers’ analyses did not change this decision. The NICE TA reported additional evidence submitted to the committee from clinical experts and concluded that, because of the higher cost of tacrolimus and pimecrolimus and the potential unknown long-term adverse effects of treatment with these products, the experts would not recommend either calcineurin inhibitor as first-line treatment.

Economic evaluations published since the NICE TA

An industry-funded Canadian study modelled the cost-effectiveness of pimecrolimus and topical corticosteroids.322 The effectiveness data came from three industry RCTs that were not referenced in the cost-effectiveness study, but they included children. It was not possible to ascertain whether they were among the RCTs considered in the guideline (the patient numbers were different to those reported in the RCTs described above). Resource use was expressed in Canadian dollars and outcomes expressed as QALYs, with QALY weightings converted from the trial IGA scores (0.99, 0.92, 0.84 and 0.74 for IGA scores 1 to 4, respectively). The study concluded that pimecrolimus was a cost-effective option given a cost-per-QALY threshold of 50,000 Canadian dollars. The study had only limited value since the costs and QALY values were derived from outside the UK and the source of the effectiveness data could not be verified.

A more recent US study published in 2006 was also industry-funded and based on clinical data from an industry trial that compared pimecrolimus 1% with conventional therapy (emollients together with topical corticosteroids for flares) for the prevention of flares for 1 year in children and young people.297 This study also used QALY weightings converted from IGA scores (0.98, 0.95, 0.88 and 0.72 for IGA scores 1 to 4, respectively). No modelling was undertaken, but the incremental cost-effectiveness ratio (ICER) of pimecrolimus versus conventional therapy was reported to be around $34,000 per QALY, concluding that it was likely to be a cost-effective option in the USA.

The results of the economic analysis suggest that topical corticosteroids could be a cost-effective option compared with pimecrolimus for mild to moderate atopic eczema on the face and body in children. The results also suggest that tacrolimus may be cost-effective compared with topical corticosteroids for more severe atopic eczema, but the results were not robust owing to the high level of uncertainty in the parameters used in the models. Because of the high cost of topical calcineurin inhibitors, more robust evidence of their effectiveness is required to determine their relative cost-effectiveness compared with other therapies.

From evidence to recommendations

Clinical trial data for topical calcineurin inhibitors published since the NICE TA was prepared provided additional evidence in support of the recommendations of the NICE TA. There was still a lack of data comparing topical calcineurin inhibitors with topical corticosteroids. The NICE guidance was adopted in this guideline.

It is the GDG’s view that the main advantage of topical calcineurin inhibitors over topical corticosteroids is that topical calcineurin inhibitors do not cause adverse effects such as skin atrophy (thinning of the skin). This is particularly beneficial when treating delicate sites such as the face, where the skin barrier is very thin and the amount of topical corticosteroid that passes through the skin can be enough to cause atrophy. It should be noted that both tacrolimus and pimecrolimus can be used on thin skin.

The GDG believes that topical calcineurin inhibitors should not be used under occlusion without specialist advice because of the risk of increased absorption.

Recommendations for topical calcineurin inhibitors (including research recommendations) are presented in Section 7.11.

7.4. Dry bandages and medicated dressings (including wet wrap therapy)

Various types of dressings can be used in the management of atopic eczema, including dry wraps, wet wraps, occlusive and semi-occlusive dressings and medicated bandages (see Table 7.3). A polythene adhesive film impregnated with fludroxycortide is also available (Haelan® Tape).

Table 7.3. Dressings used in the management of atopic eczema.

Table 7.3

Dressings used in the management of atopic eczema.

Bandaging produces occlusion leading to increased absorption of topical preparations. Other effects may also occur, including antipruritic effects, cooling and skin protection.

A survey of 233 members of the British Society of Paediatric Dermatology in 2001/2002 (40% response rate) found wide variation in UK practice in relation to how wet wrap therapies were used.323

Studies considered in this section

The HTA of treatments for atopic eczema did not cover dry bandages or medicated dressings.26 Other narrative reviews were checked for studies of any design.234,324,325 Where available, controlled trials evaluating the effectiveness of dry bandages and medicated dressings in children with atopic eczema were considered for this section. Where RCTs were not available, studies of any design were considered.

Overview of available evidence

Four RCTs249,250,326,327 evaluated the effectiveness of wet wrap dressings applied over topical corticosteroids (fluticasone, hydrocortisone and mometasone). The comparator was emollient (vehicle) in one study250 and conventional treatment (topical corticosteroids plus emollients without wet wraps) in the other three.249,326,327 The safety of topical corticosteroids under wet wrap dressings was considered in a non-randomised controlled trial328 and in three case series.329–331

A brief report on the use of fluticasone in the wet wrap method was also identified, which only included seven patients (three children). Severity (SCORAD) and cortisol levels were reported after 2 weeks’ treatment, but the report generally lacked information about the patients, their condition and other treatments used. Therefore, it was not considered further.332

Occlusive and medicated dressings

No RCTs evaluating the effectiveness of dry bandages, occlusive or medicated dressings (including silver-impregnated silk bandages or dressings) in the treatment of atopic eczema in children were identified. The use of a hydrocolloid dressing on top of clobetasol propionate lotion (no strength specified) in children and adults with refractory atopic eczema was reported in one case series (n = 48). It was not clear how many children were included in the series (the age range was 7–69 years) and no results were reported separately for children. Therefore, the study was not considered further.333 [EL = 3]

Topical corticosteroids versus vehicle under wet wrap dressings

One RCT evaluated the effects of 5 days’ inpatient treatment with wet wrap dressings of mometasone furoate 0.1% or vehicle applied twice daily in children aged 2–17 years with an exacerbation of atopic eczema. Lesions had to be present on the insides of both elbows or the backs of both knees for entry to this left–right study. Outcomes considered were disease severity (SCORAD), transepidermal water loss and S. aureus colonisation. Changes in SCORAD scores were shown only in graphs, with no numerical data provided. Improvements were evident in both groups, although this was reported to be greater in those treated with mometasone (P < 0.01). There were no significant differences between groups in transepidermal water loss. No data were shown for S. aureus colonisation.250 [EL = 1−]

Topical corticosteroids under wet wrap dressings versus conventional treatment

One RCT compared the effectiveness of hydrocortisone 1% ointment under wet wrap dressings with conventional treatment (emollient and hydrocortisone 1% ointment) in children with moderate to severe atopic eczema (SCORAD scores ≥ 15; n = 50 randomised, 45 analysed).249 Wet wrap dressings were used for 24 hours a day for 1 week, then for 12 or 24 hours a day as required for a further 3 weeks. It was not made clear whether wraps were used on the whole body. After 4 weeks’ treatment, there was no significant difference between the two groups in changes in severity (SCORAD), the quantity of hydrocortisone 1% ointment used, or in the proportion of children who used a sedating antihistamine. Reductions in SCORAD scores of 55% and 59% were reported with wet wrap versus conventional treatment. Significantly more children treated with wet wraps used antibiotics compared with conventional treatment (22% versus 0%, P = 0.05). Nurse- and carer-rated improvements were not significantly different between groups (proportions ‘much better’ or ‘better’ 65% versus 59% (nurse rating) and 70% versus 64% (carer rating)). Significantly fewer carers considered that the wet wraps were easy to use compared with conventional treatment (39% versus 73%, P = 0.036). While no children withdrew from conventional treatment, five (22%) withdrew from the wet wrap group owing to non-adherence.249 [EL = 1−]

The second RCT (a pilot study) also compared hydrocortisone 1% and emollients under wet wrap dressings with conventional treatment (emollient and hydrocortisone 1%) in children with atopic eczema affecting 30% or more of their body surface area (n = 19).326 Wet wrap dressings were applied twice daily for the first week, then only at night for the second week. Both groups used only an emollient during the third week. It was not made clear whether wraps were used on the whole body. No significant differences were found between groups in changes in SASSAD severity scores (mean change −10.3 versus −15.7, 95% CI for the difference −18 to 2), or in quality of life (IDQoL mean change −2 versus −7, 95% CI for the difference −10 to 3; DFI mean change −2 versus −5, 95% CI for the difference −14 to 2). However, this small pilot study may have been underpowered to detect differences. The study reported that the mean 2 month cost to the NHS was approximately £19 for a child under 2 years and £11 for children aged 2–15 years. Improvements in sleep were noted in both groups, but no between-group analysis was reported. Two children from each group withdrew from treatment and it was assumed these were included in the analysis. Reasons for withdrawal were folliculitis and inability to attend follow-up in the group treated with wet wraps, and non-adherence and treatment failure in the control group. In total, there were two cases of folliculitis among those treated with wet wraps.326 [EL = 1+]

One RCT considered the effectiveness of wet wrap dressings using mometasone furoate 0.1% and fluticasone propionate 0.005% ointments, both diluted to one-tenth their strengths, compared with continued treatment with the same preparations without wet wrapping.327 Children with moderate to severe refractory atopic eczema were enrolled (n = 40; 27 completed treatment and were analysed). Treatment was applied once a day over a 4 week period without wet wraps, or for 2 weeks without wet wraps followed by 2 weeks of application under wet wraps. It was not made clear whether wraps were used on the whole body. While reductions in disease severity score were noted for each group, no between-group comparisons were reported, nor were differences in baseline values accounted for. Disease extent scores fell significantly in both wet wrap groups (this outcome was not evaluated in the standard treatment group). Subjective assessment of disease impact on daily life was significantly reduced with the mometasone wet wrap, but not with the fluticasone wet wrap; again no between-group analysis was reported.327 [EL = 1−]

Studies of other designs that considered adverse effects

The first report of the wet wrap technique was published in a letter. Children aged 9 months to 16 years were treated with hydrocortisone 0.5% ointment or a 10% dilution of betamethasone valerate 0.01% under wet wraps for 2–5 days. Suppression of serum cortisol levels was evident in all children during treatment but returned to normal 2 weeks later (n = 30).334 [EL = 3]

A non-randomised controlled study focused on the effects of 2 weeks’ treatment with various dilutions of fluticasone propionate 0.05% under wet wraps on serum cortisol levels (n = 31 children aged 5 months to 13 years).328 However, data were poorly reported, with some presented only in graphs and with selective reporting of numerical data. While the authors claimed that the data suggested that weaker corticosteroid dilutions are associated with lower risk of hypothalamic–pituitary–adrenal (HPA) axis suppression, this was not evident from the data reported. Similarly, while disease severity was also measured, incomplete data were reported. Folliculitis was reported in 42%.328 [EL = 2−]

Three case series involving a total of 36 children also measured early-morning serum cortisol levels in children treated with topical corticosteroid therapy under wet wrap dressings. In the first case series, mometasone furoate 0.1% (diluted to 10% or 15% with emulsifying ointment) was applied once daily under wet wrap dressings for 2 weeks (n = 12). Early-morning plasma cortisol was measured in two-thirds of the children, with a result below the lower limit of the usual range recorded for one child. However, no baseline data were provided for comparison with this result. Folliculitis and a ‘tight sensation’ were reported as adverse events by 25% of children.330 [EL = 3]

The second case series found that SCORAD severity scores decreased significantly from baseline after 9 days’ treatment with fluticasone propionate 0.05% under wet wrap dressings in children and adults with refractory atopic eczema (n = 26, 14 children). Overall median serum cortisol levels fell significantly from baseline to day 7, but none of the values were below the lower end of the reference range (200 nmol/l).329 [EL = 3]

The third case series measured lower leg length and urinary excretion of deoxypyridinoline crosslink as markers of growth and bone turnover in children treated with wet wrap dressings (n = 8).331 Diluted beclometasone dipropionate (10% or 25%, in seven children) or emollient (one child) was applied under wet wrap dressings for 24 hours a day for 2 weeks, followed by overnight use for 1 week and then ‘as required’. After median follow-up of 12 weeks (range 2–18 weeks), lower leg length velocity rates and bone turnover did not appear to be different from baseline values.331 [EL = 3]

Evidence statement for dry bandages and medicated dressings (including wet wrap therapy)

RCTs evaluating the use of topical corticosteroids under wet wrap dressings were generally of poor quality. The results for treatment given over 2–4 weeks were conflicting, with no clear evidence of a difference in effectiveness (measured by disease severity and/or quality of life) between wet wrap and conventional treatment with topical corticosteroids plus emollients. The one RCT that compared wet wraps over topical corticosteroid versus vehicle did not provide sufficient information to enable conclusions to be drawn. [EL = 1−]

Use of wet wrap therapies was associated with higher use of antibiotics and higher withdrawal rates in one study. [EL = 1−] Folliculitis was reported in 20–42% of children across several studies. Carers found that wet wrap treatment was less easy to apply than conventional treatment. [EL = 3]

Up to 2 weeks’ use of topical corticosteroids under wet wrap dressings did not appear to affect children’s growth or bone turnover, although these data were derived from small studies. Reports of suppression of serum cortisol levels after 2–5 days’ use have been documented. [EL = 3]

There was an absence of evidence regarding the effectiveness of dry bandages, medicated and occlusive dressings for the treatment of atopic eczema in children.

Cost-effectiveness

No cost-effectiveness studies of dry bandages or medicated dressings, including wet wrap dressings, were identified.

From evidence to recommendations

The GDG found no evidence that wet wrap therapy was more effective or cost-effective than conventional treatment for mild to moderate atopic eczema, but this may reflect the power and quality of the available studies. It is the GDG’s view that wet wrap treatment with a topical corticosteroid can be beneficial in some cases, such as severe atopic eczema, very dry skin, flares that are not controlled by conventional topical corticosteroid application and limbs that are heavily scratched at night. The risk of systemic adverse effects from topical corticosteroids increases under occlusion and is proportional to the body surface area being treated. Therefore the duration of wet wrap treatment over topical corticosteroids should be limited to 7–14 days. [EL = 4]

Recommendations for dry bandages and medicated dressings (including research recommendations) are presented in Section 7.11.

7.5. Antihistamines and other antipruritics

There are a number of treatments available for severe pruritus and these include antihistamines and other antipruritics such as coal tar and bath oil preparations. Coal tar has been widely used in the past as a treatment for chronic atopic eczema. Antihistamines block the activity of histamine at receptor sites in the skin (predominantly H1 receptors), which alleviates itching and reduces the wheal and flare response, hence reducing urticaria. The relative antipruritic, anti-urticarial and sedative effects of antihistamine drugs vary.

Antihistamines are classified according to their sedative properties. Sedating antihistamines (also referred to as first-generation antihistamines) such as alimemazine (formerly known as trimeprazine), chlorphenamine (formerly known as chlorpheniramine), clemastine, cyproheptadine, hydroxyzine and promethazine act non-selectively, and tend to be shorter-acting (6–12 hours). Non-sedating antihistamines (also referred to as second-generation antihistamines) such as cetirizine, levocetirizine, loratadine, desloratadine and fexofenadine bind more selectively to peripheral histamine H1 receptors, although sedation can still occur. They have a longer duration of action (about 24 hours), except in infants where the drug may be metabolised more rapidly.

Studies considered in this section

The HTA of treatments for atopic eczema was checked for evidence relating to children.26 Where available, RCTs evaluating the effectiveness of antihistamines or other antipruritics (coal tar, bath oil preparations and/or others) in children with atopic eczema were considered for this section. Where RCTs were not available, studies of any design were considered.

Overview of available evidence

RCTs evaluating the use of cetirizine, chlorphenamine, clemastine, cyproheptadine, hydroxyzine, ketotifen and loratadine in children with atopic eczema were identified. No trials of any design considered the effects of preparations containing coal tar on pruritus. One study compared the effects of two different coal tar 1% preparations in individuals (mostly children) with atopic eczema, but in terms of global improvement and patient preference rather than pruritus.335 An RCT compared the effectiveness of an aqueous lotion containing sodium cromoglicate 4% with placebo in children with atopic eczema.336 A study that investigated the use of a non-proprietary preparation of sodium cromoglicate used specifically for the study was not considered to be relevant to UK clinical practice and was not considered further.337

Antihistamines for the treatment of pruritus associated with atopic eczema

Cetirizine

One double-blind placebo-controlled randomised trial considered the effectiveness of cetirizine in the treatment of mild to moderate pruritus in children aged 6–12 years with atopic eczema (n = 22).338 The dosage of cetirizine given was dependent on body weight: 5 mg/kg daily was given to those weighing 30 kg or less, and 10 mg/kg daily to those over 30 kg. After 8 weeks’ treatment there were significant differences between the two groups in terms of clearance of all signs and symptoms of atopic eczema (73% cetirizine versus 18% placebo, P < 0.02), and in use of concomitant therapy (disodium cromoglicate or topical corticosteroids: 18% cetirizine versus 82% placebo, P < 0.01). Severity of pruritus and erythema was also measured in the study, but no numerical results were reported.338 [EL = 1+]

Chlorphenamine

One double-blind RCT compared the effectiveness of chlorphenamine with placebo in children aged 1–12 years who had nocturnal itching and scratching associated with atopic eczema (n = 151).339 Treatment was given for 4 weeks. The dosage of chlorphenamine given was 1 mg once daily for children aged 1–5 years and 2 mg once daily for children aged 6–12 years. Where itching was not reduced by the initial dose, a second identical dose was permitted from 3 hours after administration of the first dose. If itching had not improved at the end of the first 2 weeks of treatment then the dosage was doubled (2 mg and 4 mg for children aged 1–5 years and 6–12 years, respectively). Use of emollients and hydrocortisone 1% was permitted during the trial.339

After 4 weeks’ treatment, no significant differences were identified between groups in any outcome. Severity of itching (graded on a five-point scale) was not significantly different between the two treatment groups: 56% in both groups had no itching, and 33% from the chlorphenamine group versus 29% from the placebo group reported minimal itching (P = 0.745). There was no significant difference between groups in terms of other outcomes assessed (investigator’s rating of intensity of signs and symptoms, quantities of emollients or hydrocortisone use). Overall, 13% reported a total of 29 separate non-serious adverse events; no further details were reported.339 [EL = 1+]

Hydroxyzine versus cyproheptadine

One double-blind RCT evaluated the effects of hydroxyzine and cyproheptadine on pruritus (day and night) in children aged 2–16 years (mean age approximately 8 years) with an acute exacerbation of atopic eczema (n = 20).340 The doses taken were 1.25 mg/kg three times daily (t.d.s.) of hydroxyzine (up to 30 mg t.d.s.), and 0.25 mg/kg t.d.s. cyproheptadine (up to 6 mg t.d.s.). The doses used were higher than those generally used in UK practice. The children were also using an emollient preparation three times daily, but no other medications were permitted. Improvement in both day and night pruritus was significantly greater with hydroxyzine than cyproheptadine after 7 days’ treatment (mean improvement in daytime pruritus 32% versus 6%, P < 0.001; nighttime pruritus 48% versus 30%, P < 0.005). Physician-rated improvement of the severity of the condition at endpoint was also significantly greater in the hydroxyzine group. Other than sedation, noted in two children in the hydroxyzine group and three in the cyproheptadine group, no adverse effects were reported.340 [EL = 1+]

Clemastine versus ketotifen

A double-blind RCT compared the effectiveness of clemastine and ketotifen in children (mean age 9 years) with atopic eczema (n = 284 randomised, 255 analysed).341 After 4 weeks’ treatment, the proportion of children whose condition was moderately improved based on the investigator’s rating was significantly higher with ketotifen; no other differences in the other six ratings were noted. In terms of individual symptoms, improvement in itching, erythema/papule and excoriation/scratch was found in significantly more children treated with ketotifen (itching 79% versus 57%, erythema/papule 73% versus 58%, excoriation/scratch 70% versus 54%). Other than the proportions reporting adverse events, which were similar in the two groups (clemastine 13% versus ketotifen 10%), details of adverse events were lacking.341 [EL = 1−]

Loratadine versus placebo

A study evaluating the use of loratadine in conjunction with topical mometasone furoate 1% cream in children with atopic eczema was identified (n = 50). Although the volume (and not strength) was reported in the paper, it was assumed that the only available proprietary preparation of loratadine was used (5 mg/5 ml). The dose given was 5 ml for children who weighed up to 30 kg, and 10 ml for those weighing more than 30 kg. After 15 days’ treatment, there were no significant differences between groups in any outcome (improvement in severity (SCORAD) scores, physician’s assessment of global improvement or pruritus score). Dizziness was reported by one child in each group; there were no reports of drowsiness or difficulty in awakening.342 [EL = 2+]

Antihistamines used preventatively in children with atopic eczema

The Early Treatment of the Atopic Child (ETAC) study considered whether cetirizine could prevent the onset of asthma, and also provided longer term safety data for cetirizine. In this double-blind RCT, 18 months’ treatment with cetirizine was compared with placebo in children aged 12–24 months with active atopic eczema (n = 795).343–346 The dosage of cetirizine given was 0.25 mg/kg twice daily. Both groups were permitted to use topical or systemic medication if required.

There was no difference in cumulative prevalence of asthma between active and placebo groups after 18 months of treatment (38%). The proportion of children who reported one or more episode of urticaria was significantly lower with cetirizine (5.8%) than placebo (16.2%, P < 0.001).343,346 There were no significant differences between the two treatment groups in the proportions who used topical preparations, or in the duration of their use (emollients, corticosteroids, nonsteroidal anti-inflammatory creams, ‘tar’, antibiotic/antiseptics) or systemic oral antibiotics. The quantities of other medications taken were not reported.

In the subgroup of children with more severe atopic eczema (SCORAD score of 25 or more; 44%), the mean percentage days’ use of moderate to potent topical corticosteroids was significantly lower in the cetirizine group (25.8% cetirizine versus 35.1% placebo, representing 51 fewer days’ use of such topical corticosteroids during the total trial period, P = 0.014). Despite the difference in topical corticosteroid use, overall reduction in disease severity scores (SCORAD) did not differ significantly between groups (change from baseline −39% cetirizine versus −37% placebo).343 The proportion of children in the cetirizine group who were given other oral antihistamines was significantly lower than in the placebo group (18.6% versus 24.9%, P = 0.03). The mean percentage days of their use was also reported to be statistically significantly lower in the cetirizine group compared with placebo (3.4% versus 4.4%, P = 0.035), although the difference of 5 days over 18 months was not clinically important.343 [EL = 1++]

There were no significant differences between the cetirizine and placebo groups in terms of rates of serious symptoms and adverse events (9.3% cetirizine versus 13.6% placebo, P = 0.053), or hospitalisation (9% cetirizine versus 11.8% placebo, P = 0.189). Similarly, there were no significant differences between groups in neurological symptoms or events, including insomnia, fatigue, somnolence, hyperkinesis, nervousness, emotional lability or febrile convulsions. Electrocardiogram and laboratory test results in both groups were within normal limits.344

Assessments of behaviour and psychomotor development were undertaken in some children (41% and 20%, respectively). There were no significant differences between groups in mean scores on the behavioural screening questionnaire or in psychomotor development scores measured by the McCarthy test.344,345 [EL = 1++]

Sodium cromoglicate

A double-blind RCT compared the effectiveness of an aqueous lotion containing sodium cromoglicate 4% with placebo (lotion base only) in children with moderate to severe atopic eczema (n = 114, age range 2–12 years).336 Treatment with emollients and topical corticosteroids continued as usual during the study. After 12 weeks, the severity of the atopic eczema (SCORAD score) was reduced by 36% (13.2) using sodium cromoglicate 4% compared with 20% (7.6) using placebo (mean difference 5.6, 95% CI 1.0 to 10.3). Further analysis comparing clinically relevant treatment success (defined as a reduction in SCORAD of at least 25% without an accompanying increase in topical corticosteroid use) showed that treatment with sodium cromoglicate 4% and placebo was associated with treatment success in 50% and 30% of children, respectively (OR 2.29, 95% CI 1.06 to 4.94). [EL = 1+]

Evidence statement for antihistamines and other antipruritics

Controlled trials evaluating antihistamines and other antipruritics for atopic eczema in children were few in number and generally evaluated short-term use (1–8 weeks’ treatment) in relatively small numbers of children. The indications for treatment with an antihistamine were not always made clear. One large and longer study showed no difference in the use of topical corticosteroids between children taking cetirizine and those taking placebo. Where antihistamines were used to treat itching associated with atopic eczema in children, the available data were conflicting; there was no evidence that cetirizine or chlorphenamine led to greater improvements in pruritus compared with placebo. There was some evidence from one small trial that hydroxyzine was more effective than cyproheptadine in relieving pruritus over a period of 1 week. [EL = 1+] The RCT comparing ketotifen and clemastine was of poor quality which did not allow conclusions to be drawn. [EL = 1−] None of the studies considered the impact of antihistamine treatment on the children’s or families’ sleep or quality of life. No studies evaluated the use of sedating antihistamines for sleep disturbance in children with atopic eczema.

Cetirizine was as well tolerated as placebo in an 18 month trial evaluating its use for the prevention of asthma in young children with atopic eczema. In children with more severe atopic eczema (SCORAD ≥ 25), cetirizine reduced the use of moderately potent and potent topical corticosteroids. [EL = 1++]

Details of adverse effects were generally lacking across the studies that evaluated antihistamines for the treatment of atopic eczema, although none reported clinically important differences between antihistamines and placebo groups.

An RCT showed that treatment with an aqueous lotion containing sodium cromoglicate 4% was more effective than placebo (lotion base only) in reducing the severity of atopic eczema in children. [EL = 1+]

Cost-effectiveness

No published economic evaluation studies were identified. Antihistamines are not expensive treatments; some have shown some beneficial effects in treating atopic eczema in children, and these prescriptions are likely to be cost-effective. Sedating and non-sedating antihistamines cost about £5 to £10 per month (excluding outliers) (BNFC 2007).232 Although no economic analysis was reported, the likelihood is that this is a cost-effective treatment in the circumstances for which it is recommended.

The ETAC trial343 showed that children with atopic eczema given cetirizine used less topical corticosteroid and had a lower rate of urticaria than those treated with placebo. The reduction in treatment costs (not having to treat urticaria) may well have offset the initial (low) cost of the antihistamine, but the study did not report this. Without overall quality of life information, it was not possible to evaluate whether any additional cost of treatment was offset by the reduced costs and increased quality of life in reducing rates of urticaria.

From evidence to recommendations

The GDG’s view was that antihistamines can be helpful in some circumstances (for example, when treating children with atopic eczema involving an element of urticaria), and that these treatments are likely to be cost-effective. Although the evidence base was poor, clinical experience still supports short-term (7–14 days) use of sedating antihistamines in children with atopic eczema that causes debilitating sleep disturbance to them or their families or carers. If treatment with sedating antihistamines is successful, it can be repeated during flares if needed. Non-sedating antihistamines can be used for 1 month in the first instance if there is severe itching or urticaria. This trial can be continued if successful but should be reviewed every 3 months. The GDG noted that all antihistamines can alter mood and cognitive function and long-term use should be avoided.

One RCT has shown that treatment with an aqueous lotion containing sodium cromoglicate 4% is effective in reducing the severity of atopic eczema in children. However, the treatment is not licensed for use in the UK and the GDG has, therefore, not recommended its use.

Recommendations for antihistamines and other antipruritics (including research recommendations) are presented in Section 7.11.

7.6. Treatment for infections associated with atopic eczema

Bacterial and viral infections that occur secondarily to atopic eczema, and their signs and symptoms, are:

Damage to the epidermal skin barrier from inflammation and scratching allows bacterial colonisation, particularly with S. aureus, which represents about 90% of the total aerobic bacterial flora of people with atopic eczema; this compares to 30% in unaffected individuals.26 Heavy colonisation of the skin with S. aureus has been reported in people with atopic eczema even when the skin is not clinically infected, and this may contribute to continuing disease activity.347–349 The density of S. aureus tends to increase with the clinical severity of atopic eczema lesions.350–354

Serous exudate encourages bacterial growth and frequently leads to clinical infection (impetiginised eczema). This is associated with increased inflammation, heavy yellowish crusting and sometimes pustules and even frank blisters of impetigo, which can spread rapidly. The role of S. aureus in non-clinically infected atopic eczema skin or borderline infection is far from clear.26

In people with atopic eczema, a high rate (73%) of self-contamination from S. aureus carrier sites (nose, subungual spaces (under the nails), axillae (armpits), groin and the periauricular area (ears)) or from colonised skin lesions has been described.355–357 Bacterial transmission between children with atopic eczema and family members has also been reported.358–361

Staphylococcus aureus can produce enterotoxins (enterotoxins A–E and toxic shock syndrome toxin-1).362 These cause a number of diseases, some of which may be followed by fever and shock. The toxins can act as superantigens interacting with immune cells to induce or enhance inflammation of the skin (and other sites).363 There is some evidence to suggest that the density of S. aureus is more important that the presence of superantigens in aggravating atopic eczema lesions.364 Superantigens can also induce glucocorticosteroid insensitivity, which may increase the severity of atopic eczema.362

Severe atopic eczema associated with severe recurrent infections, especially deep abscesses or pneumonia, should be investigated as it may be associated with rare diseases such as Job syndrome, Netherton syndrome, Wiskott–Aldrich syndrome, and selective IgA deficiency.

Increased infection rates are associated with the use of immunosuppressive agents (such as corticosteroids) for the management of atopic eczema.

Eczema herpeticum (Kaposi’s varicelliform eruption) is a generalised vesicular eruption caused by the herpes simplex (cold sore) virus (usually type 1). It is relatively uncommon, considering that both atopic eczema and recurrent herpes simplex occur in about 20% of the population. It has been suggested that children with atopic eczema are no more likely to acquire herpes simplex infections than are children unaffected by atopic eczema.365 However, another study reported that adults who had had atopic eczema between the ages of 0 and 14 years had a greater incidence of recurrent herpes simplex infections than did non-atopic controls.366

Other viral infections such as varicella (chicken pox) may occasionally be very widespread in atopic eczema, mimicking eczema herpeticum.367

While infection with other organisms such as viral warts, including molluscum contagiosum, was once thought to be more common in people with atopic eczema, there is no evidence to support this.368 Such organisms may, however, be more widespread or persistent because of scratching and/or the use of immunosuppressive therapies such as topical corticosteroids and topical calcineurin inhibitors.

Pityrosporum ovale and tinea (ringworm) infections are no more common in children with atopic eczema than other children.369,370 Using topical corticosteroids can alter the clinical appearance of tinea infections, allowing low-grade spread of the causal fungus.

Yeast fungi (mainly Candida spp. and Rhodotorula spp.) are thought to be present on the skin of approximately 40% of people with atopic eczema. They are difficult to eliminate and can aggravate the course of the disease.371

Other itchy skin conditions such as scabies (infestation with Sarcoptes scabiei var. hominis) may coexist or be confused with atopic eczema. Scabies worsens the usual itching associated with atopic eczema and this usually results in considerable impetiginisation, which can mask the signs of scabies.

7.6.1. Identification of infections

Studies considered in this section

Most of the literature on skin infection in association with atopic eczema relates to S. aureus, although other microorganisms are associated with infected atopic eczematous skin. The studies considered in this section describe bacterial infections (n = 14) and viral infections (n = 28). No relevant studies were identified for Pityrosporum ovale, tinea, yeast fungi or scabies infections.

Overview of available evidence

Bacterial infections
Staphylococcus aureus alone

Staphylococcus aureus infection associated with atopic eczema was described in one case series372 and eight case reports of extremely rare complications caused by S. aureus.373–376 [EL = 3]

The case series reported 22 secondary infections (31.4%) with S. aureus in 57 children with atopic eczema (severity mild to severe) aged 4 months to 14 years followed for an average of 4.73 months.372 [EL = 3]

Four of the case reports described children under 12 years of age with severe atopic eczema and a confirmed S. aureus infection. All children exhibited pustules in the affected areas and one child had pustules and impetigo.373 Two of the case reports described S. aureus septicaemia as a complication of infected atopic eczema in an infant and a 3-year-old child.374 One case report described S. aureus-induced osteomyelitis associated with cutaneous colonisation of S. aureus in 4-year-old boy.375 The third case report described a 3-year-old boy with severe atopic eczema and history of recurrent skin infections who was admitted to hospital with skin sepsis. Acute bacterial endocarditis was diagnosed as a result of S. aureus infection. Following treatment for his condition, the boy had two further episodes of septicaemia due to Proteus mirabilis and Pseudomonas aeruginosa.376

Staphylococcus aureus with streptococcus species

Staphylococcus aureus complicated with streptococcal infections and atopic eczema were described in three case series377–379 and one case report.380 [EL = 3]

The first case series reported on 190 children (aged 7 weeks to 17 years, median age 3 years) with atopic eczema (no details of severity were reported) attending a hospital clinic and studied for a mean of 13 months.377 [EL = 3] Seventy-six children (40%) had exacerbations of atopic eczema due to bacterial infections and in 52 (32%) infection recurred within 3 months. Twenty-five cases (15%) led to hospital admission. Staphylococcus aureus was recovered in 97% of cases and in combination with beta-haemolytic streptococci in 62%.

The second case series described 174 cases of streptococcal impetigo associated with atopic eczema, of which 112 were in children under the age of 14 years.378 [EL = 3] The most frequent infectious agents were group A streptococci (71% Streptococcus pyogenes) followed by group G (19.5%) and group B (9.8%) Streptococcus agalactiae. Streptococci were isolated as sole pathogens in 28% of cases and in the remaining cases they were co-infecting with S. aureus.

In the third case series, six of 36 children under the age of 12 years with atopic eczema were found to have lesions infected with streptococci in addition to S. aureus.379 [EL = 3] There were two cases of Streptococcus pyogenes, three cases of streptococcus group G, one of which also involved Streptococcus agalactiae, and one other unidentified streptococcus.

Two further case reports describing unusual infections were not considered to be relevant to the clinical management of atopic eczema.380,381 [EL = 3]

Viral infections
Eczema herpeticum

Eczema herpeticum was described in six case series382–387 and nine case reports.388–396 [EL = 3]

Eczema herpeticum may arise in normal-looking skin without evidence of active atopic eczema and sometimes in people who have not had active atopic eczema for many years. Lesions are all at the same stage of evolution. They start as small, grouped, circular blisters which often show a central depression (umbilication). They are all remarkably similar in size and appearance but quickly become eroded and crusted and often confluent in some areas. Transmission is by direct contact with infected secretions. The severity of eczema herpeticum ranges from localised disease to widespread dissemination and very rarely herpetic encephalitis and death. Mortality rates for untreated eczema herpeticum have been reported to be 6–10%.387 The cause of death, though not always clear, may have been an undetected immune deficiency state such as Wiskott–Aldrich syndrome or a secondary bacterial infection with S. aureus and streptococcus species. Repeated attacks do occasionally occur and should prompt a search for underlying immune deficiency.

Varicella

Infection with varicella (chicken pox) was described in one case control study.397 [EL = 2−]

In otherwise healthy children with varicella infections, systemic symptoms are usually mild and complications are rare. In immunologically compromised children and children on steroid therapy, the infection is more likely to be associated with an extensive eruption, high fever, pneumonia and life-threatening complications.397 In a case–control study comparing 32 children with atopic eczema and a varicella infection with 34 children with a varicella infection but no atopic eczema 37.5% (controls 5.9%) had persistent fever, 31% (5.9%) had profuse eruptions, and 87.5% (17.6%) had severe pruritus. 397 [EL = 2]

Viral warts

Viral warts were described in one case–control study.368 [EL = 2+] Infection with verrucae vulgaris was described in one case report.398 [EL = 3] Viral warts have traditionally been thought to be more common in children with atopic eczema than unaffected children. However, no evidence was identified to support this. In fact, one case–control study reported that warts were noted less frequently in children with atopic eczema than unaffected children at age 11 years and 16 years. 368 [EL = 2+]

Molluscum contagiosum

Molluscum contagiosum was described in two case series399,400 [EL = 3] and eight case reports.401–403 [EL = 3] No evidence was identified to suggest that molluscum contagiosum was any more common in children with atopic eczema than in other children.

7.6.2. Antimicrobial agents

Treatments for infected atopic eczema involve the use of systemic antibiotics active against S. aureus, topically applied antibiotics, and antiseptic agents applied directly to the skin or mixed with emollients applied directly to the skin or bath additives.

Antibiotics are important for treating overt secondary bacterial infections in children with atopic eczema. Flucloxacillin is useful for treating S. aureus infections although oral preparations are often considered unpalatable by children. Phenoxymethylpenicillin is used for Streptococcus pyogenes. Erythromycin is used when there is resistance to flucloxacillin or in patients with a penicillin allergy, although it is associated with nausea.404 Side effects present less commonly with clarithromycin compared with erythromycin. Clarithromycin and erythromycin have equivalent antibacterial activity. In cases of penicillin allergy, there is a 6–10% risk of allergy to cephalosporins.

Studies investigating antimicrobial agents for atopic eczema considered reduction of skin colonisation by microbes as an outcome, as well as effectiveness in treating overt clinical infection. Reduction of S. aureus colonisation on the skin of children with atopic eczema using oral antibiotics (erythromycin, cloxacillin, flucloxacillin, cefuroxime axetil), topical antiseptics (chlorhexidine, potassium permanganate, an antibacterial soap (triclosan 1.5%)), acid-electrolyte water therapy and antibacterial silk clothing have been described.405–411

Contamination of topical treatment agents with microorganisms such as S. aureus, Pseudomonas aeruginosa and Alternaria alternata has been reported, although not in conjunction with cases of atopic eczema in children.412–415

Studies considered in this section

Six studies were identified in relation to treatment of infection associated with atopic eczema in children. Antibacterial treatment of infected atopic eczema in children was described in two RCTs,241,416 [EL = 1−] one cohort study417 [EL = 2−] and one case report.418 [EL = 3] A topical steroid/antibiotic combination treatment was described in one controlled double-blind within-person (left–right body comparison) study which combined data from children and adults.419 [EL = 2−] Two case series reported the use of antimicrobial emollient preparations.240,243 [EL = 3]

No studies were identified that evaluated the effectiveness of treatments for streptococcal infections, nor for antiseptics, topical antibiotics, antiseptic–topical corticosteroid combinations or antivirals as treatments for infected atopic eczema in children.

Overview of available evidence

Antimicrobial emollient preparations

A double-blind RCT compared a bath emollient containing benzalkonium chloride and triclosan with the regular bath emollient (Oilatum Plus® versus Oilatum®). All the children had atopic eczema displaying features of recurrent infection and/or frequent exacerbations (n = 30 randomised, 26 analysed). After two treatment periods of 4 weeks, some improvement in total clinical score (signs and symptoms plus area affected) were reported from baseline, although no baseline scores were reported. It was, therefore, difficult to quantify the benefit. It was also reported that there were no significant differences between groups in global change or impression scales or in self-reported severity of the condition, but no numerical data were presented. Pruritus was reported in 23% of children overall.241 [EL = 1−]

A case series reported the use of an emollient containing antimicrobials (benzalkonium chloride and chlorhexidine hydrochloride (Dermol 500 lotion)) in children. The children were receiving treatment for eczema (whether the eczema was atopic was not reported) and in need of emollients to manage their dry skin condition. Between 81% and 87% reported that dryness and itching of the face/neck and limbs/trunk were better after 2 weeks’ treatment and satisfaction rates were also high. No adverse effects were reported during the trial (n = 39).240 [EL = 3]

A publication consisting of seven case reports of irritant reactions to a bath oil preparation containing the antimicrobials benzalkonium chloride and triclosan (Oilatum Plus) was also identified. Four of the seven were children aged up to 12 years who had infected atopic eczema. In two children who used the preparation as directed, reactions consisted of an erythematous rash that developed immediately and dry non-pruritic desquamation after 2 weeks’ use. In the other two children, quantities of bath oil in excess of that recommended were used; the adverse effects were described as ‘an irritant reaction’ affecting the skin flexures which developed over several months, and erythema and scaling around the mouth and on the trunk (in the second case subsequent use of the same product at the correct concentration was well tolerated).243 [EL = 3]

Antibacterials

In the other RCT, 30 children with suspected S. aureus superinfected atopic eczema (age range 6 months to 12 years) were randomised to either oral cefadroxil (50 mg/kg/day) in two doses or placebo for 2 weeks.416 [EL = 1−] Twenty-eight of the 30 children had superinfections with S. aureus alone or in combination with group A streptococci as diagnosed by swab culture. After 2 weeks, all children on the active treatment were infection free compared with nine out of 17 in the placebo group. Severity of atopic eczema improved in both active and placebo groups, but there were no statistically significant differences between groups. Physician-rated global assessment was significantly in favour of the active treatment (P = 0.009), although patient-rated global assessment was similar in both groups.

In the cohort study, 35 children (ages 2–11 months) with atopic eczema and methicillin-resistant S. aureus (MRSA) infection were treated with oral nadifloxacin (15–30 g) and bufexamac ointment (a nonsteroidal anti-inflammatory; 20–40 g) or with bufexamac ointment alone for 4 weeks.417 [EL = 2−] After 4 weeks, MRSA infections were absent in the active treatment group and continued to be so for the next 3 months, serum IgE levels were significantly reduced (P < 0.001) and severity of atopic eczema using a simple inflammation score was significantly improved (P < 0.0001). In contrast, the control group showed no resolution in MRSA infection and no changes in IgE serum levels or severity of atopic eczema.

A case report describing a 4-year-old boy with atopic eczema and an MRSA infection who developed osteomyelitis in the fingers was considered to be a rarity and not important to clinical management of eczema.418 [EL = 3]

Topical corticosteroid and antibiotic combination treatment

In one controlled double-blind within-person (left–right body comparison) study, 81 dermatology patients, of whom 26 were children (median age 9 years, range 1–15 years), were treated with betamethasone 17-valerate 0.1% and fusidic acid 2% cream on one side of their body and betamethasone 17-valerate 0.1% alone on the other side for 1 week.419 [EL = 2−] Sixty of the 81 patients were diagnosed as having atopic eczema (no details of severity or individual data for children were reported), and the majority of patients were judged clinically to have a degree of impetiginised dermatosis. Although all patients improved within the week of treatment, there were no significant differences in clinical improvement or reduction of bacterial colonisation between the two treatments. Patient preference tended towards the combination treatment.

7.6.3. Antimicrobial resistance

With the emergence of S. aureus strains with antibiotic resistance to agents such as methicillin and, more recently, fusidic acid, prolonged use of any antibiotic will sooner or later be associated with the emergence and increased prevalence of resistant strains.

It is important to distinguish between laboratory-tested antibiotic resistance of microorganisms and that of microbes on colonised or infected skin. Use of topical antibiotics results in high localised concentrations of antibiotics that can override laboratory resistance and produce a clinical response.

Studies considered in this section

Five studies were identified that evaluated antimicrobial susceptibility of infections associated with atopic eczema: three were case control studies (one in children only and two that combined data from children and adults),420–422 [EL = 2−] one a case series involving children only,423 [EL = 3] and one a survey.424 [EL = 3] Adult studies were considered because of the lack of evidence from children.

In one case–control study the bacterial flora of 50 children with atopic eczema (mean age 4.4 years) was determined on their first admission to hospital and compared with that of 20 control children.420 [EL = 2−] Bacterial colonisation was more prevalent in the children with atopic eczema compared with control children. Staphylococcus aureus was the most common pathogen: 32% were phage group II (that is, strains not associated with impetigo or streptococcal scalded skin syndrome) and the density of S. aureus was proportional to the severity of the atopic eczema. Resistance to penicillin was present in 88% of strains and to two or more antibiotics in 38% of the strains.

Bacterial skin colonisation in another case control study involving 33 children and adults (age range 3 months to 32 years, mean age 12.7 years) with mainly mild to moderate atopic eczema was compared with a control group.421 [EL = 2−] There was greater colonisation with S. aureus in people with atopic eczema compared with controls (42% versus 5%, P = 0.003) and this was related to severity of the atopic eczema. All S. aureus isolated from people with atopic eczema were sensitive to cloxacillin, cefalexin, clindamycin and co-trimoxazole; 92% were sensitive to erythromycin, but only 13% were sensitive to penicillin and ampicillin.

One case–control study investigated 48 children and adults (age range 6 months to 75 years, mean age 6.7 years) of whom 48% had atopic eczema (no details of severity were reported).421 [EL2−] Seventy-eight percent of S. aureus isolated from people with atopic eczema were resistant to fusidic acid compared with 9.6% in non-dermatology patients; 96% of people with atopic eczema who had S. aureus resistant to fusidic acid had used a preparation containing fusidic acid in the previous 6 months.

One case series described antimicrobial susceptibility of S. aureus in 115 children (mean age 2.7 years) with moderate to severe atopic eczema.423 [EL = 3] Staphylococcus aureus was isolated from 87% of children. Antimicrobial susceptibility testing showed resistance to erythromycin in 18% of cases, to roxithromycin in 19%, to fusidic acid in 6% (resistant or ‘intermediately susceptible’), to amoxicillin 13% and to clindamycin in 1%. All strains isolated were susceptible to oxacillin, amoxicillin/clavulanic acid, cefadroxil and cefuroxime.

A 5 year retrospective study of the characterisation and susceptibility to fusidic acid of S. aureus in the Carmarthen area suggested an increased incidence of fusidic acid resistance particularly with paediatric patients with infected eczema and impetigo.424 [EL = 3] In children aged 10 years or younger (n = 255, including some children with atopic eczema), fusidic acid resistance increased from 5.1% to 24.6% between 1999 and 2001. Over the same period, prescriptions of fusidic acid preparations increased in general practice, although they remained constant in hospital pharmacies.

Evidence statement for infections associated with atopic eczema in children

The majority of children with atopic eczema have skin colonised with Staphylococcus aureus. A high rate of self-contamination from S. aureus carrier sites (nose, nails, axillae, groin and ears or from colonised skin) has been reported. [EL = 3] Where children developed overt signs of clinical infection this was usually due to S. aureus, although streptococcus species (principally Streptococcus pyogenes) were sometimes involved. Mixed infections of S. aureus and streptococci have also been reported. [EL = 3] Other types of bacterial infection that occur in association with atopic eczema are rare and generally thought not to be any more common in children with atopic eczema than in other children. Infection with herpes simplex (eczema herpeticum), varicella (chicken pox), molluscum contagiosum, human papillomavirus, Pityrosporum ovale, tinea (ringworm), yeast fungi and scabies have been documented. [EL = 3] Eczema herpeticum can be life-threatening. Varicella may exacerbate atopic eczema or present as widespread varicella resembling eczema herpeticum. No evidence was found on how to treat varicella in children with atopic eczema.

The evidence for the effectiveness of antibiotic treatments for infected atopic eczema was lacking, with the few studies identified being of poor quality. [EL = 3] The available studies provided some evidence for the effectiveness of antimicrobials, but evidence for cost-effectiveness was lacking. Irritant effects due to inappropriate or long-term use of antimicrobial emollients have been reported in a number of clinical studies.

Contamination of emollient preparations with S. aureus and Pseudomonas aeruginosa has been reported. [EL = 3]

There was evidence for increasing prevalence of resistance of microorganisms to antibiotic agents (such as fusidic acid, flucloxacillin and erythromycin) using in vitro tests on bacteria cultured from skin swabs of children with atopic eczema. [EL = 3]

Although there were isolated case reports of extremely rare complications of infection associated with atopic eczema, the GDG considered these to have little relevance to routine clinical practice. [EL = 3]

Cost-effectiveness

No health economics issues were identified in relation to which clinically significant infections occur secondarily to atopic eczema in children, or in relation to the signs and symptoms of such infections. Assessment should take place within routine clinical consultations and requires no additional healthcare resources. Erythromycin is as effective as clarithromycin and less costly (£2.35 for a 28-tab pack of erythromycin versus £5.39 for a 14-tab pack of clarithromycin, BNFC 2007),232 but no studies were identified that considered the cost-effectiveness of treatment for infected atopic eczema in children.

From evidence to recommendations

Colonisation of the skin with bacteria (mainly Staphylococcus aureus) and overt clinical infection are both associated with an increase in severity of atopic eczema, although there is a lack of agreement as to the density at which the presence of bacterial colonisation exacerbates atopic eczema.

Eczema herpeticum (due to herpes simplex virus) is under-recognised and, if not diagnosed promptly, the child’s condition may deteriorate rapidly. Eczema herpeticum should, therefore, be an indication for urgent referral. Varicella may exacerbate atopic eczema and present as widespread varicella resembling eczema herpeticum or lead to secondary impetiginisation. Molluscum contagiosum can be more extensive in children with atopic eczema than in other children because of spread from scratching, and it often seems to worsen atopic eczema locally at the site of lesions.

The GDG believes that, owing to the potential dangers of herpes simplex virus and eczema herpeticum, treatment should be started with oral aciclovir at the first suspicion of herpes simplex virus in a child with atopic eczema in order to control the infection and prevent the development of eczema herpeticum. If eczema herpeticum is suspected, oral or intravenous aciclovir can be given depending on the clinical situation.

Some oral antibiotics are unpalatable, but in many cases there is no alternative. The GDG’s view was that flucloxacillin should normally be the first-line treatment for S. aureus and streptococcal infection because it is active against both. Erythromycin should be used when there is local resistance to flucloxacillin and in children with a penicillin allergy because it is as effective as cephalosporins and less costly. However, erythromycin is associated with nausea. Side effects present less commonly with clarithromycin compared with erythromycin. The GDG’s collective experience suggested that in cases of penicillin allergy there is a 6–10% risk of allergy to cephalosporins.

It is the view of the GDG that topical antibiotics, including those combined with topical corticosteroids, should be used to treat localised overt infection only, and for no longer than two weeks.

The GDG believes that healthcare professionals should refer to local guidelines for advice on local patterns of resistance to antimicrobials and such patterns should be reviewed regularly.

Skin swabs taken for bacteriological culture are generally of limited use owing to the universal colonisation of skin with S. aureus in people with atopic eczema. Skin swabs can, however, be useful where there is recurrent infection, concern about antimicrobial resistance to antibiotics commonly used for S. aureus infection or clinical suspicion of unusual organisms.

The GDG believes that antiseptics such as triclosan and chlorhexidine can be used as an adjunct therapy for decreasing bacterial load. Some antiseptics can be irritant and very occasionally cause contact allergic dermatitis so they should only be used at appropriate dilutions and for short periods of time.

There is potential for re-infection when products in open containers contaminated with S. aureus and Pseudomonas aeruginosa are used.

Recommendations for infections associated with atopic eczema in children (including research recommendations) are presented in Section 7.11.

7.7. Stepped approach to management

Evidence relating to the definition, identification and management of flares of atopic eczema in children, management and monitoring between flares (maintenance therapy), and optimal combinations and/or sequences for using different treatments were sought for this section.

7.7.1. Identification and management of flares

Atopic eczema is usually episodic, with the episodes being called flares (factors that might precipitate flares were described in Chapter 6 and treatments for infections that might accompany flares were described in Section 7.6). There is no universally accepted definition of a flare. The question of what is a flare has been addressed in a systematic review.425 [EL = 1+] The review identified 15 studies that provided definitions, all of which were clinical trials of interventions to treat atopic eczema in children and/or adults (some of which are considered elsewhere in this guideline). The definitions for flare or relapse used were:

  • a change in severity score above a set threshold (change in SCORAD score of 50–80% or more than 15 points; increase in TIS score of at least 4 points; increase of 70% in Costa’s SSS score; or increases of more than 75% in disease activity scores) – seven studies
  • a composite of an IGA score of at least 4 and topical corticosteroid use for 3 days following a 7 day period free of topical corticosteroid use – three studies
  • the need to use topical corticosteroids (or systemic treatment in one study) – three studies
  • an IGA score of at least 3 with a score of 2 or 3 for any two signs or symptoms (erythema, itch, papulation and induration/oedema) – one study
  • a scratch score of more than 2 on a five-point scale for 3 consecutive days – one study.

The ISOLATE study, which involved children and adults from eight countries including the UK, reported disease characteristics during a flare (n = 2002, 39% were parents of children aged 2–13 years).96 Flare was defined as ‘a sudden worsening of symptoms requiring a physician consultation or application of prescription medication.’ Children aged 2–13 years experienced a mean number of 8.7 flares per year, each lasting a mean duration of 14 days, thereby spending 33% of the year experiencing a flare of atopic eczema.

Although topical corticosteroids and topical calcineurin inhibitors have been widely used for the treatment of flares, little evidence was identified regarding their use specifically for this indication. The identified data consisted of:

The RCTs comparing fluticasone propionate cream 0.05% with the two hydrocortisone preparations reported improvements in all groups but greater improvement in total eczema score and in itch and sleep disturbance with fluticasone. This study was described in detail in Section 7.2.255 [EL = 1+]

The RCT that considered the use of wet wrap dressings with mometasone furoate 0.1% or vehicle in children with an exacerbation of atopic eczema was described in detail in Section 7.4.250 [EL = 1−]

The three RCTs that evaluated the use of pimecrolimus at the first sign or symptom of atopic eczema in order to prevent progression to a flare (IGA score of at least 4 and topical corticosteroids used for 3 days following a 7 day period free of topical corticosteroid use) were described in Section 7.3.297,302–304 These studies found that the proportion of children whose condition progressed to a flare was significantly lower in children who were treated with pimecrolimus compared with vehicle (both used with emollients).

A non-randomised controlled study evaluated the effects of wearing silk clothing with antibacterial properties compared with continued use of cotton clothing in children with a flare of atopic eczema (n = 46, age range 4 months to 10 years).426 ‘Flare’ was not defined. All children applied emollients but the use of topical corticosteroids was not permitted. After a follow-up period of 1 week, SCORAD severity scores had reduced significantly from baseline in the silk clothing group (30% reduction, P = 0.003), but not in the control group (2% reduction, P = 0.886). No between-group analysis or baseline data were reported. Therefore it was not possible to determine whether groups were similar other than in the intervention being evaluated. Among children wearing silk clothes, a significant reduction in severity (SCORAD) was reported for areas covered by silk clothes compared with similar uncovered areas in the same child (reductions of 42%, P = 0.001 versus 16%, P = 0.112).426 [EL = 2−]

7.7.2. Management and monitoring between flares

Three double-blind RCTs considered the effectiveness of topical fluticasone propionate for reducing relapse of atopic eczema, one involving children and adults427 and two involving adults only.428,429 The control group in each study was the vehicle base of the topical corticosteroid preparation. Emollients were also used daily.

The studies involving adults were considered because of the relative lack of data regarding maintenance therapy in children aged 0–12 years. The first RCT evaluated fluticasone propionate 0.05% cream in children and adults with moderate to severe atopic eczema (n = 348, 66% aged 2–17 years).427 Atopic eczema had been stabilised by up to 4 weeks’ treatment with fluticasone propionate 0.05% cream applied twice daily before randomisation to receive a reduced dose of fluticasone or vehicle (once-daily use 4 days a week for 4 weeks, followed by once-daily use twice a week for 16 weeks). Relapse was defined as an IGA score of 3 or more (scale 0–5), and a score of 2 or 3 (on a scale of 0–3) for any of three signs or symptoms (erythema, pruritus and papulation/induration/oedema). The relapse rates were 66% in the vehicle group, whereas in children using fluticasone propionate 0.05% cream they were 27% (OR 8.1, 95% CI 4.3 to 15.2, P < 0.001). The median time to relapse was 5.1 weeks in the vehicle group but could not be quantified in the fluticasone group because most were controlled at the end of the follow-up period. Individual adverse effects were not reported, although it was stated that the incidence of these did not differ significantly between groups. None of the children or adults had ‘evidence of skin atrophy’ (not defined). Of 44 cosyntropin stimulation tests undertaken (it was not stated whether they were undertaken in children or adults), two did not reach the required post-stimulation serum cortisol level of at least 18 μg/dl (the levels were 9 μg/dl and 17 μg/dl).427 [EL = 1+]

Two other double-blind RCTs evaluated the use of fluticasone to treat flares and to prevent subsequent relapses of atopic eczema, but in people aged 12 years and over. The first study consisted of two treatment periods: initial treatment of the flare with one of four fluticasone options, then, following stabilisation, patients either continued with a fluticasone option or received treatment with vehicle base only (n = 376).428 [EL = 1+] Patients were randomised to the whole treatment sequence at the outset. A flare was defined as a score of 4 or more on TIS.

The four options used in the initial treatment of the flare were fluticasone 0.05% cream applied once daily or twice daily, and fluticasone 0.005% ointment applied once daily or twice daily. Following the stabilisation period of up to 4 weeks, treatment with fluticasone cream or ointment, or its vehicle base, was applied for up to 16 weeks. During this time, the frequency of application was reduced to twice weekly on two consecutive evenings. The risk of relapse was significantly lower in those treated with fluticasone propionate 0.05% cream or 0.005% ointment compared with vehicle (hazard ratio (HR) 5.8, 95% CI 3.1 to 10.8, P < 0.001 with fluticasone 0.05% cream versus vehicle; HR 1.9, 95% CI 1.2 to 3.2, P = 0.01 with fluticasone 0.005% ointment versus vehicle). Median time to relapse was longer than 16 weeks (the duration of the study) with both fluticasone preparations, compared with 6.1 weeks in both vehicle groups. Adverse events noted during the stabilisation phase were three reports of visual signs of skin atrophy (two having telangiectasia and striae and one having telangiectasia).428

The second study involving adults also reported a lower relapse rate in those treated on two consecutive days per week with fluticasone propionate 0.005% ointment compared with vehicle for 16 weeks (n = 54).429 However, it was not possible to tell from the data reported whether groups were similar at baseline in parameters other than the intervention. [EL = 1−]

7.7.3. Combining treatments

When considering how to combine treatments for atopic eczema in children the GDG aimed to evaluate:

  • the effectiveness and cost-effectiveness of combination products (for example, a topical corticosteroid with an antimicrobial versus either alone) – see Section 7.6
  • the effectiveness and cost-effectiveness of treatments used in combination (for example, topical corticosteroids alongside emollients) versus one of the treatments used alone
  • the effectiveness and cost-effectiveness of different treatment strategies (for example, short-term use of a potent topical corticosteroid versus longer term use of a less potent preparation, or topical corticosteroids compared with topical calcineurin inhibitors for the management of flares)
  • how to sequence treatments for optimal effect (that is, the effective and cost-effective use of available treatments) including which treatments to use in specific circumstances (considering severity, signs and symptoms, health-related quality of life and other criteria affecting quality of life).

There was a lack of evidence for how to combine or sequence treatments for atopic eczema. There were few trials of true treatment alternatives (for example, topical corticosteroids compared with topical calcineurin inhibitors), and thus it was not possible to establish an optimal sequence of treatments in terms of clinical effectiveness data alone.

An RCT that considered different strategies for using topical corticosteroids was described in Section 7.2.254

Some of the trials of antihistamines reported that they were used in conjunction with a topical corticosteroid and emollient, but the comparison in these trials was only placebo.339,342 [EL = 2+]

Similarly, in studies evaluating topical calcineurin inhibitors, emollients were used in all treatment arms. The reporting of whether emollients were also used in studies involving topical corticosteroids was generally poor.

Evidence statement for stepped approach to management

In clinical trials, a flare has been defined in a variety of ways, predominantly involving severity or IGA. A minority of studies defined a flare in terms of the need to use certain additional treatments, which does not inform when to use these treatments. There was no published consensus on how to define or identify a flare.

There were some data showing that topical corticosteroids are effective when used specifically to treat a flare. [EL = 1+] RCTs showed that pimecrolimus 1% cream reduced the progression to flare compared with vehicle when used at the first sign or symptom of atopic eczema. [EL = 1+] No conclusions could be drawn from one small study of poor quality that considered the use of silk versus cotton clothing for 1 week in children who experienced a flare of atopic eczema. [EL = 2−] When used following the stabilisation of a flare, maintenance treatment with fluticasone propionate (0.05% cream or 0.005% ointment) applied twice weekly for 16–20 weeks was more effective than its vehicle base in reducing the relapse rate in children. [EL = 1+]

No evidence to evaluate the optimal combination or sequence of treatments for atopic eczema in children was identified.

Cost-effectiveness

There was a lack of evidence of the effectiveness of combinations of treatment and consequently there was no evidence of the cost-effectiveness of these treatments. Economic evaluation requires treatment outcomes to be evaluated using the same units to allow direct comparison of the costs and health benefits of treatment alternatives. These data were not available and therefore it was not possible for the GDG to reach any meaningful consensus as to the likely comparative advantage of one combination of treatments over another.

From evidence to recommendations

In the absence of consistent definitions for what constitutes a flare, the GDG’s view was that in clinical practice a flare should be defined as an increase in clinical severity (redness, oedema (swelling) or itching) of the condition. Parents usually recognise when a child’s atopic eczema is flaring because it becomes more itchy and red and the child scratches more, thus the child will be complaining or showing that their skin is causing a problem over and above what they would normally expect.

The GDG believes that it is important to try to identify what is precipitating a flare because this will influence the treatment choice or intervention. Additionally, it is important to recognise a flare early because early treatment prevents damage to the skin barrier which results from the itch–scratch cycle. In the GDG’s view, treating dry skin, which can be an early sign of a flare, with an emollient may prevent worsening of a flare.

The data regarding prevention of flares in adults are probably only relevant to older children with chronic established atopic eczema which is constant; the data may not be transferable to younger children with complete clearance between flares and who might thus be using the topical corticosteroid unnecessarily.

In the absence of published evidence regarding optimal strategies for combining or sequencing treatments for atopic eczema in children, the GDG’s consensus was that treatment should follow a stepped approach, taking into account the severity of and degree of control of the atopic eczema, possible trigger factors and the effect on quality of life and psychosocial wellbeing of the child and their family/caregivers. Emollients should always be used as minimal maintenance therapy, and their use should be continued during flares. One or more of the following treatments should be used in addition to emollients during flares: topical corticosteroids, topical calcineurin inhibitors, dry bandages or medicated dressings (including wet wraps), antihistamines, appropriate treatment for infected eczema, and, in some severe cases, phototherapy and systemic treatments (see Section 7.8). For mild atopic eczema, treatment options are emollients and mild topical corticosteroids; for moderate eczema (excluding the face and neck), emollients can be used with moderate topical corticosteroids, tacrolimus or bandages; for severe eczema (excluding the face and neck), treatments include emollients, potent topical corticosteroids, tacrolimus, bandages, phototherapy and systemic treatments. Treatment should be stepped up or down according to severity and clinical response.

Recommendations for stepped approach to management (including research recommendations) are presented in Section 7.11.

7.8. Phototherapy and systemic treatments

This section covers phototherapy and treatments given orally or by injection that modulate the immune response.

Studies considered in this section

The HTA of treatments for atopic eczema was checked for evidence regarding phototherapy or systemic immunomodulators in children with atopic eczema.26 Where available, RCTs evaluating the effectiveness of these interventions in children with atopic eczema were considered in this section. Where RCTs were not available, studies of any design were considered.

7.8.1. Phototherapy

Phototherapy involves exposure to ultraviolet light (UVA or UVB rays) under controlled conditions. Psoralen, a photoactive drug, can be given with UVA (known as PUVA) to enhance the effectiveness of phototherapy. The mechanism of action of phototherapy in atopic eczema is not completely understood but is believed to involve immunosuppression.26 The wavelength of UVB phototherapy is 290–320 nm, narrow-band UVB 311–313 nm, and UVA 320–400 nm. The risks of developing skin cancers following long-term damage to the skin by ultraviolet light are well known and these relate to high lifetime cumulative dosages. The risk associated with one course of phototherapy is thought to be extremely low to negligible.430

Overview of available evidence

Studies reporting the use of phototherapy using UVB (including narrow-band), UVA, and PUVA in the treatment of atopic eczema in children were identified.

Narrow-band UVB

The use of pimecrolimus 1% cream in combination with narrow-band UVB irradiation was evaluated in a 6 week RCT in children and young people (n = 26, aged 5–17 years).431 [EL = 1−] No other treatments (including emollients) were allowed during the study. The two treatment arms were as follows: pimecrolimus applied to the whole body and irradiation to one half; and pimecrolimus applied to half the body and irradiation to the whole body. Within-patient comparisons were reported for each treatment arm, which found no significant difference in improvements in EASI scores (score reductions of 53–59%). Changes in pruritus scores were also similar in all patients. Two patients reported intractable generalised pruritus and tender erythema.431 [EL = 1−]

A cohort study aimed to compare the effects of narrow-band UVB irradiation on the skin flora of children with atopic eczema and vitiligo (n = 20, mean age 9.5 years).432 [EL = 2−] The amount of UVB exposure was the same in both groups although no details of the regimen or duration of follow-up were reported. Levels of cutaneous aerobes, anaerobes, staphylococci (including Staphylococcus aureus) fell; the changes were reported to be statistically significant (P < 0.05), but it was not clear whether this was from baseline or between groups (or both). SCORAD scores fell significantly from baseline in children with atopic eczema. Adverse effects of treatment were not considered.432 [EL = 2−]

Three case series described the use of various phototherapy regimens in children with a range of skin conditions and reported data for children with atopic eczema separately.433–435

The first case series described the use of UVB given three times a week for 7–20 weeks, mean 15 weeks (n = 20, aged 16 months to 11 years, 25% with atopic eczema).433 The number of treatments given ranged from 20–61 (mean 41). Outcomes were reported vaguely, with all children ‘moderately improved’ (not defined). Burning and erythema necessitating the temporary discontinuation of treatment was reported in two children.433

The second case series reported the outcomes of combined UVA and UVB treatment, given three or five times per week for an unknown duration (n = 53, aged 4–16 years, 40% with atopic eczema).434 Reduction in SCORAD score of at least 90% was reported in 45%, reduction of 70–90% in 23%, and reduction of 50–70% in 32%. Four people experienced mild erythema.434 [EL = 3]

The third case series described the outcomes of narrow-band UVB phototherapy (n = 77, aged 4–16 years, 32% with atopic eczema).435 Details of the treatment regimen (frequency of photo-therapy and its duration) were lacking, as were demographic details. Of the children with atopic eczema, 68% had minimal residual disease at the end of treatment. Adverse effects (total group) included erythema (30%), anxiety (6.5%), and infection with herpes simplex (2.6%) or varicella zoster (1.3%).435 [EL = 3]

The use of narrow-band UVB in children and young people with atopic eczema was also described in a letter (n = 40, aged 2.5–15 years).436 Details of the frequency of phototherapy and duration of treatment were again lacking. It was reported that 23% had an excellent response (not defined), 58% a good response, and 20% a poor response (treatment discontinued). Longer term follow-up data for 24 of the 32 patients who completed treatment were reported, which showed relapse rates of 20% within 6 weeks, 50% at 3–4 months, and 25% at 6–9 months; the remaining patient was in remission at 2 years. Adverse effects reported were facial erythema in 35%, xerosis in 25%, herpes labialis in 5% and burning in 2.5%.

One case series437 described all paediatric patients with severe atopic eczema who had undergone narrow-band UVB phototherapy between 1999 and 2005 in a particular clinic (n = 60, aged 4–16 years). Adverse events were experienced by 14 children. These included well-demarcated erythema, painful erythema and reactivation of herpes simplex virus. Follow-up data were incomplete and there was no comparator group. [EL = 3]

PUVA

Two case series described the response to PUVA therapy. In the first, PUVA was given twice or three times a week in children with severe atopic eczema (n = 53, aged 6–16 years).438 After a mean of 9 weeks’ treatment, 74% achieved at least 90% clearance of their eczema. The remainder did not have clearance or ‘near’ clearance; most withdrew from treatment. Overall, 38% also received oral prednisolone during the early phase of treatment, which was then gradually tapered off. The cumulative dose of UVA and the number of irradiation treatments in children also receiving prednisolone was lower. At 1 year, 69% remained in remission. Adverse effects reported were the development of freckles (30%), blistering (19%), recurrent herpes simplex (9%) and acute exacerbations of asthma (4%). It was reported that there was no evidence of corneal or lens opacities, and that liver function tests remained normal.

In the second case series, children and young people (aged 10–14 years) were treated with PUVA, for an unknown duration (n = 15).439 Clearance or near clearance was achieved in all except one person who withdrew from the study because of intolerance to the heat of the irradiation cabinet. Short courses of oral prednisolone were also taken by one-third of patients when it was not possible to increase the dose of UVA irradiation owing to skin irritability. Time to remission ranged from 0.3 to 1.8 years (median 1 year), and duration of remission from 0.25 to 4.2 years (median 1.1 year). Adverse effects reported were freckles (20%) and cutaneous herpes simplex and photo-onycholysis (7% each).439 [EL = 3]

Cost-effectiveness

No evidence was identified regarding the cost-effectiveness of systemic immunomodulators or phototherapy for the treatment of atopic eczema in children.

7.8.2. Systemic treatments

Overview of available evidence

Studies reporting effectiveness data for ciclosporin, azathioprine, systemic corticosteroids, interferon gamma and intravenous immunoglobulin in the treatment of atopic eczema in children were identified. It should be noted that interferon gamma is not available as a treatment for atopic eczema in the UK. Most available data related to ciclosporin. There were limited numbers of RCTs, with most data being reported as small case series or case reports for all the treatments considered. No studies evaluating the use of methotrexate, mycophenolate mofetil or systemic tacrolimus in children with atopic eczema were identified.

Ciclosporin

The studies identified for ciclosporin in children with atopic eczema consisted of one RCT,440 four case series,64,441–445 and four publications describing one or more cases.446–449 Only children with atopic eczema who had failed to respond to other treatments were included in these studies.

The RCT of ciclosporin use in children compared two treatment strategies – a 3 month course and 12 months’ continuous use, both at a dose of 5 mg/kg/day (n = 43 randomised, 40 analysed, age range 2–16 years).440 [EL = 1−] No significant differences were reported between groups in any outcome (severity (SASSAD), or body surface area affected) at 1 year. More than half (57%) of those in the 3 month group were treated continuously or had extended treatment periods. Quality of life was also assessed, but the method used and results obtained were not reported. Adverse effects occurring in at least 5% of each treatment group were nausea, paraesthesia, hypertrichosis, swollen gums, headache, rhinitis, upper respiratory tract infection, abdominal pain, folliculitis and hyperuricaemia.440

In the first case series, the response (not defined) to ciclosporin therapy was ‘good’ or ‘excellent’ in 89% (median duration 6 weeks; n = 18, aged 3–16 years). The initial dose used was 5–6 mg/kg, thereafter the dose was titrated according to response. The relapse interval (relapse defined as the requirement for potent topical corticosteroids or further systemic treatment) was a median of 6 weeks (range 0–38 weeks). One child experienced nausea, but otherwise there were no adverse effects. There were no significant changes in serum creatinine or in blood pressure.441 [EL = 3]

In a case series of children treated with ciclosporin 5 mg/kg for 6 weeks, significant improvements were reported in all outcomes (severity (SASSAD), extent, pruritus, sleep disturbance, irritability, reduction in topical corticosteroid use; n = 27, aged 2–16 years). However, results were reported only in graphs with no numerical changes reported. Significant improvements in quality of life were also reported, although the measurement tool used was not specified. In terms of global response and tolerability, more than 75% reported at least considerable improvement in symptoms, and at least 92% reported good or very good tolerability (the child’s/parent’s and investigator’s assessments gave similar results). The most common adverse effects were headaches (26%), abdominal pain (22%) and nausea (15%). There were no statistically or clinically significant changes in serum creatinine levels or in blood pressure. There was one case of a transient increase in serum bilirubin levels which normalised (treatment was not discontinued).64 [EL = 3]

In another case series, children with severe atopic eczema were treated with ciclosporin 2.5 mg/kg/day which could be increased to 5 mg/kg/day (n = 10, aged 22–189 months).442–444 After 8 weeks’ treatment SCORAD scores had reduced by 35% or more in nine children (the reduction was 32% in the remaining child). Seven of the nine children’s atopic eczema did not relapse during the additional 4 week follow-up period. There were no cases of hypertension and no significant changes in serum creatinine levels. Serum bilirubin levels increased by 2.5 μmol/l, the increase being statistically significant. Tolerability was regarded as good or excellent in nine children by their own or their parents’ assessment and in eight children by the investigator’s assessment.442 The quality of life of the mothers of these children was also assessed. Of the five subscales of the German FEN quality of life assessment tool, there were significant improvements in the psychosomatic wellbeing and the emotional coping of the children’s mothers.443 [EL = 3]

In the fourth case series, children aged 2–16 years with severe atopic eczema were treated with ciclosporin 2.5–5 mg/kg/day for 8 weeks. The SCORAD score fell significantly from baseline (P < 0.001). Greater effectiveness was reported in children only colonised with S. aureus compared with those clinically infected with S. aureus (mean SCORAD scores were lower, P < 0.01). Other data were only reported in graphs. A significant reduction in S. aureus density was seen in colonised but not infected children.445 [EL = 3]

Other identified information regarding the use of ciclosporin consisted of case reports containing varying amounts of detail (not providing case history, or only noting dosages used, or reporting specific adverse effects).

One publication described three children aged 2, 4 and 5 years who had been treated successfully with ciclosporin 5 mg/kg/day for 8 weeks without any adverse effects. Relapse occurred once treatment stopped, but after varying intervals.446

Another case report described a change in formulation of ciclosporin in a child aged 2.5 years. Treatment was switched from one formulation (Sandimmun®; oral form no longer available in the UK) after 6 weeks of therapy to another formulation (a microemulsion, Neoral®, currently the only oral formulation of ciclosporin available in the UK). Treatment was changed because of deterioration in the child’s atopic eczema. After 8 weeks’ treatment with the microemulsion, the investigator-rated severity score reduced by 55%, and itching, sleep and irritability all improved by 37–47% (rated by mother).447 [EL = 3]

In one case report, reduction in raised blood pressure was seen during treatment with ciclosporin 5 mg/kg/day in a 6-year-old boy with severe atopic eczema, asthma and hay fever. The raised blood pressure at baseline was believed to be due to stress related to atopic eczema, sleep deprivation or previous/concurrent treatment, which included potent topical corticosteroids, inhaled corticosteroids and ‘occasional’ oral prednisolone. Thus the normalisation of blood pressure was considered to be due to successful management of the condition with ciclosporin.448 [EL = 3]

One publication reported two cases of raised alkaline phosphatase levels in children aged 2 years who were treated with ciclosporin. The levels normalised after treatment withdrawal.449 [EL = 3]

Systemic corticosteroids

A crossover double-blind RCT compared 4 weeks’ treatment with oral plus nasal beclometasone dipropionate with placebo in children (n = 27, aged 3–14 years) with atopic eczema.450 The oral beclometasone used was the contents of capsules for inhalation mixed with some water; the inhaled product was a proprietary nasal spray. Significantly greater improvements in redness, surface damage and lichenification were seen with beclometasone compared with placebo. The daytime itch score and use of antihistamines were significantly lower in the systemic corticosteroid group, while sleep loss scores and daily use of topical corticosteroids were not significantly different between groups. Parental global assessment indicated that children fell into the ‘no change’ to ‘somewhat better’ category, but the difference between groups was statistically significant, the children treated with beclometasone tending towards ‘somewhat better’. No adverse effects were reported during treatment.450 [EL = 1−]

Other isolated reports of the use of systemic corticosteroids for atopic eczema in children were identified, but only vague details were provided in the reports. A small case series reported the effectiveness of a 3 day course of intravenous methylprednisolone 20 mg/kg/day in children with severe atopic eczema and raised serum IgE levels in whom conventional treatment had failed (n = 7, aged 3–14 years). Improvements in severity were reported in five of the children (reduction in a generic score from a mean of 49 to less than 8), which persisted for a mean of 10 months (range 3–18 months). The other two children only experienced mild and transient improvement. IgE levels were ‘unaffected’ by therapy (no further details reported). Adverse effects were not considered.451 [EL = 3]

The successful use of oral prednisone (5 mg daily) in a 7-year-old child with atopic eczema in whom standard treatment (including topical corticosteroids and emollients) had failed was documented.452 [EL = 3]

Another publication reported the worsening of atopic eczema in two children (aged 6 and 8 years) on withdrawal of a systemic corticosteroid (the drug was not specified).453 [EL = 3]

Azathioprine

One case series described the use of azathioprine 2.0–3.5 mg/kg/day to treat severe atopic eczema in children who had normal thiopurine methyltransferase levels (n = 48, aged 3–16 years).454 The total duration of treatment was 983 months in the whole group but the range and mean/median duration of treatment and/or follow-up were not quoted. (Thiopurine methyltransferase is an enzyme that metabolises azathioprine, and it is believed that those with low levels are at higher risk of developing myelosuppression from the drug). Based on parental global assessment of the child’s condition at 3 months, 58% had an excellent response (at least 90% improvement) and 27% had a good response (60–90% improvement), while the remaining 15% were classified as having an inadequate response (less than 60%). Overall, 48% were also treated with prednisolone at some time during azathioprine treatment. Adverse effects during treatment were one case each of eczema herpeticum, gastrointestinal symptoms (nausea, vomiting, diarrhoea) and a possible hypersensitivity reaction (manifested as urticaria and vomiting). There were no cases of neutropenia. Other transient effects were abnormalities of liver function tests (10%), lymphopenia (31%), and thrombocytopenia (2%).454 [EL = 3]

In another publication, the same investigators described azathioprine treatment in two children (aged 7 and 14 years) who had low thiopurine methyltransferase levels (below the normal range). The 14-year-old was treated with 1.25 mg/kg/day for 10 months. The 7-year-old was treated with 1 mg/kg/day for 8 months. Improvement in the atopic eczema was seen after 2 weeks’ azathioprine treatment (more than 90% in one, and ‘almost clear’ in the other). The 7-year-old had a varicella zoster infection during treatment which was treated successfully. Benefit was reported to be sustained for 8–10 months (no further details were given for beyond this period) and oral corticosteroid therapy was withdrawn in both cases.455 [EL = 3]

Methotrexate

No studies evaluating the use of methotrexate to treat atopic eczema in children were identified. Two case series in adults with moderate to severe atopic eczema reported improvements in the majority of patients treated for a median or fixed duration of 3 months (total n = 32).456,457 Methotrexate was given by intramuscular injection or orally in one study456 and orally in the other.457 In both studies, treatment was given or taken once weekly. Adverse effects reported included nausea and transient increases in liver enzymes.

Interferon gamma

One placebo-controlled double-blind RCT,458 an associated long-term follow-up study459 and five case series or case reports460–464 described the use of interferon gamma to treat atopic eczema.

The RCT included children and adults (age range 3–65 years), with some data reported separately for those aged 3–20 years (n = 83, 25% aged 3–20 years).458 [EL = 1+] However, the relative proportion of people aged 3–20 years differed between groups, with six treated with interferon gamma and 15 treated with placebo.

Interferon gamma 50 μg/m2/day by subcutaneous injection was self-administered by patients (or carers in the case of children, presumably) for 12 weeks. At the end of treatment, the proportions reporting at least 50% improvement were significantly higher in the interferon gamma than the placebo group (45% versus 21%, P = 0.016 based on the investigator’s assessment, and 53% versus 21%, P = 0.002 based on the patient’s or carer’s assessment). In those aged 3–20 years, the patient/carer ratings were 67% versus 20% (investigator’s assessment was not reported). Of six signs or symptoms evaluated, significantly greater improvement was reported with interferon gamma than placebo for erythema and excoriations, but there were no significant differences between groups for the other four parameters (pruritus, induration, dryness and lichenification). The quantity of topical corticosteroid used (triamcinolone acetonide 0.1%) was not significantly different between groups. Adverse effects reported were headaches (60% interferon gamma versus 28% placebo, P = 0.004), myalgia and chills (30% interferon gamma, not reported for placebo), transient granulocytopenia (12.5% versus 2.5%), and mild transient increases in liver transaminase levels (16.3% versus 2%).458

Twenty-four patients (aged 11–57 years) from the RCT were treated with interferon gamma for 1 year, and 16 patients for 2 years.459 [EL = 3] Reasons for discontinuation between years 1 and 2 were inconvenience and non-adherence (two each), and improvement without therapy, ineffectiveness, flu-like symptoms and unknown reasons (one each). Significant improvements in most outcomes were reported at both year 1 and year 2 (total body surface area affected, global assessment, total clinical severity and individual parameters (erythema, excoriations, pruritus, induration, dryness and lichenification)). Improvements in the associated atopic symptoms allergic conjunctivitis and rhinitis were also significant, but not asthma. No significant changes in serum IgE levels were reported. Increases in the liver enzymes aspartate aminotransferase and alanine aminotransferase were evident at year 1 and fell towards baseline at year 2. Serum creatinine was mildly elevated at year 2 but remained within the normal range. Adverse effects reported were ‘transaminitis’ (16%), headache, malaise, acne vulgaris, neutropenia, arthralgia (8% each), fever/chills, gastric and oesophageal ulcers, splenomegaly, herpes zoster, molluscum contagiosum, respiratory ‘congestion’, theophylline toxicity and postherpetic neuralgia (4% (n = 1) each).459

A second case series including children and adults (aged 3.6–57 years) reported the effects of interferon gamma therapy for atopic eczema (n = 15, 60% aged under 16 years).460 [EL = 3] Treatment with interferon gamma was for a minimum of 22 months (range 22–76 months, median 36 months), at a dose of 50 μg/m2 daily for 12 months, reduced to every other day thereafter if less than 10% of body surface area was affected on two consecutive visits. Treatment was discontinued if less than 10% of body surface area was affected on two consecutive visits on the alternate day regimen. The results showed a reduction in both total body surface area affected and in total severity score over time. Growth charts used to monitor patients aged under 16 years did not appear to show any effects on growth during the study. Treatment-related adverse effects were headaches (47%), fever (13%) and chills (6.7% (n = 1)).460

The third case series aimed to evaluate immunological parameters as predictors of success of interferon gamma therapy in patients with severe atopic eczema that had not responded to topical corticosteroids and antihistamines (n = 68, age range not reported).461 The dose used was 2 × 106 IU/m2 for 5 days in the first week, three times a week for 3 weeks, and then twice a week for another 2 weeks. Some severity data were also reported, with more than 20% (mean 63%) reduction in severity in 34%, less than 20% (mean 8%) in 44%, and no response in the remainder (22%). Adverse effects were not considered.461

The other three publications documented the use of interferon gamma in a total of ten children.462–464 The first publication reported that treatment in a 2-year-old boy was unsuccessful and was changed to interferon alpha, after which clearance of atopic eczema lesions was seen following 6 months’ treatment. The severity of the condition reduced in a 5-year-old treated with interferon gamma three times a week for 20 weeks.462

The second publication documented a lack of response in a 4-year-old boy and a 5-year-old girl. Both children had previously been treated unsuccessfully with topical corticosteroids.463

The third publication discussed the histories of children in whom the authors used interferon gamma as a last resort, all initially treated as hospital inpatients.464 The children had severe atopic eczema and other conditions or problems. However, the outcome of interferon treatment was not described clearly: it seemed that in two children treatment was successful, in one it was not, and no information was given regarding the outcomes of the other three.464 [EL = 3]

Intravenous immunoglobulin

One narrative review described literature identified in relation to the use of intravenous (IV) immunoglobulin in children with atopic eczema, which consisted of three publications.465 In four children, IV immunoglobulin was used to treat Kawasaki syndrome or idiopathic thrombocytopenia purpura, in which improvement (‘remission’) of their coexisting atopic eczema was noted within 7 days. A case report of an 8-month-old boy treated for thrombocytopenia did not find improvement of his atopic eczema. The third publication reported improvement in ‘skin score’ and in levels of cytokines (including interleukin and interferon levels) in five children with atopic eczema who were treated with IV immunoglobulin.465 [EL = 3]

Mycobacterium vaccae

One double-blind RCT evaluated the effects of killed Mycobacterium vaccae on atopic eczema in children with moderate to severe disease (n = 166, 93% completed and analysed, aged 5–16 years).63 [EL = 1−] At 12 or 24 weeks following a single intradermal injection of the preparation (either 1 mg or 0.1 mg, or placebo), there were no significant differences between groups in any outcome (severity (SASSAD), body surface area affected, patient’s global assessment, pruritus, sleep, topical corticosteroid use, or quality of life (CDLQI)). Overall, 19% had injection-site reactions (induration and erythema), and 13% had atopic eczema that was believed to be due to the injection given (32% reported atopic eczema as an adverse effect overall).

Evidence statement for phototherapy and systemic treatments

One RCT of poor quality reported no significant difference between 6 weeks’ treatment with pimecrolimus 1% cream alone or pimecrolimus 1% cream in combination with narrow-band UVB. [EL = 1−] Case series describing other phototherapy regimens in children with atopic eczema were also identified (UVB, UVA plus UVB, narrow-band UVB and PUVA), but reporting of the actual regimens used and of outcomes was generally poor. Some benefit, variously defined, was noted for a proportion of patients. Adverse effects reported include erythema, burning, blistering, dryness and the development of freckles. [EL = 3]

There was some evidence for the effectiveness of ciclosporin, systemic corticosteroids, azathioprine, interferon gamma and intravenous immunoglobulin for the treatment of atopic eczema in children, but no evidence of its cost-effectiveness. No evidence evaluating the clinical or cost-effectiveness of methotrexate or of mycophenolate in children was identified.

One RCT found no significant difference between a 3 month and a 12 month course of ciclosporin therapy in children in terms of severity or body surface area affected. [EL = 1−] Case series reported a response in the majority of those treated with ciclosporin, although the outcomes measured and the level of detail given for outcomes were lacking. Adverse effects reported included headaches, nausea and abdominal pain. None of the studies reported significant changes in blood pressure or in serum creatinine levels. [EL = 3]

A short-term crossover study of beclometasone given orally and by inhalation reported greater improvements in itch, redness, surface damage and lichenification compared with placebo, but no significant difference for sleep loss or daily topical corticosteroid use. Global assessment indicated only small benefit. [EL = 1−] Other isolated reports of systemic corticosteroid use mainly reported some response, although there were also reports of unsuccessful treatment outcomes and withdrawal effects. [EL = 3]

Case series of azathioprine use (48% of whom were also treated with systemic prednisolone at some time during treatment) reported response in the majority at 3 months. [EL = 3]

One double-blind RCT in children and adults found that significantly more patients treated with interferon gamma than placebo had a 50% or greater response at 3 months. Two of six signs/symptoms were significantly improved, with no significant difference between the groups in changes in the other four. Longer term use (up to 2 years) in some of the patients treated (aged 11 years and above) indicate sustained benefit. Other case series indicated improvements in severity and in total body surface area affected, while case reports noted both success and failure of interferon gamma treatment.

Some reports of response to IV immunoglobulin were identified in the literature, when used to treat atopic eczema, or indirectly when the intervention was used to treat another condition.

No evidence regarding the cost-effectiveness of systemic treatments or phototherapy for the treatment of atopic eczema in children was identified.

From evidence to recommendations

Phototherapy and systemic treatments have only limited evidence of effectiveness for some children with severe atopic eczema and have potentially serious adverse effects. The GDG believes that phototherapy should be considered before systemic treatments unless there are contraindications such as very fair skin or family history of skin malignancies. Phototherapy and systemic treatments should only be offered under close supervision by specialists experienced and trained in their use as they require close monitoring for safety aspects. After weighing up the benefit and harm of treatment and the costs (drug and equipment costs and specialist time), the GDG took the view that phototherapy and systemic treatments should be used only in severe cases of atopic eczema in children where other management options have failed or are not appropriate, and where the atopic eczema has a significant impact on quality of life. It is the GDG’s view that assessment and documentation of severity and quality of life should always be undertaken prior to initiating treatment with systemic treatments or phototherapy.

Recommendations for phototherapy and systemic treatments (including research recommendations) are presented in Section 7.11.

7.9. Complementary therapies

Complementary therapies are defined as a group of therapeutic and diagnostic disciplines that exist largely outside the institutions where conventional healthcare is taught and provided. These therapies can be used alongside conventional care, as the term ‘complementary therapies’ implies. Patients may also choose to use complementary therapies instead of mainstream medicine (that is, as ‘alternative therapies’). Complementary therapies have become more widely used over the past two decades, but many practitioners/practices in the UK are largely unregulated. In 2000, a report on complementary and alternative therapies by the House of Lords Select Committee on Science and Technology recommended that ‘in order to protect the public, professions with more than one regulatory body should make a concerted effort to bring their various bodies together and to develop a clear professional structure.’ In 2005, the Department of Health published a consultation document regarding the statutory regulation of herbal medicine and acupuncture and the Department is in the process of setting up a stakeholder working group to move towards regulation of these two professions.

Until recently, the majority of over-the-counter herbal medicines were classified and sold as food supplements, with little control over their quality and contents. New EU regulations regarding the regulation of herbal medicinal products came into force in the UK on 31 October 2005 to address this situation.466 Section 12(1) of the Medicines Act 1968 that allows herbal practitioners to make up personal prescriptions is also being considered for reform regarding the preparation of herbal mixtures by a third party. It is proposed that any third party producing herbal products must be able to prove good manufacturing practice.

The use of complementary therapies in children with atopic eczema and their parents/guardians was surveyed in a secondary care setting in Leicester.467 [EL = 3] The mean age of the children was 7.3 years (range 0.6–17.1 years) and ethnic origin was 59% white, 35% Indian, 3% Afro-Caribbean and 3% mixed race. Forty-six of the 100 children/parents questioned had used, or were currently using, complementary therapies. Of the 54 who had not yet used complementary therapies, 31% said they intended to try this in the future. The most commonly used therapies were Chinese herbal medicine (43%), herbal medicine (41%) and homeopathy (35%). Of the 74 episodes of treatment experienced by the users, in 26 of the incidents the child/parent felt that their atopic eczema had improved, while 39 reported that there was no change; in the remaining nine incidents the child/parent reported the eczema had deteriorated. There was a strong association between the use of complementary therapies and ethnicity. Fifty-four percent of users did so because their conventional treatment was not working, with 17% saying they were worried about side effects of conventional treatment. Thirty-nine percent of all children/parents felt that complementary therapies were safer than conventional medication although only 14% thought they were more effective. Fifty-one percent were happy to combine both types of treatment.

In another UK survey involving 80 children with atopic eczema (mean age 3.9 years), 34 (43%) had used at least one form of complementary medicine for their condition, of which herbal medicine (41%) and homeopathy (24%) were the most popular. Of these children, 44% expressed some improvement (most commonly reduction in itch), while 10% experienced deterioration in their atopic eczema.468 [EL = 3]

Studies considered in this section

The HTA of treatments for atopic eczema was checked for evidence relating to complementary therapies.26 Where available, RCTs evaluating the effectiveness of complementary therapies in children with atopic eczema were considered for this section. Where RCTs were not available, or were too short in duration to consider adverse effects, observational studies of any design were considered.

Overview of available evidence

Studies evaluating the following complementary therapies in children with atopic eczema were identified:

  • homeopathy
  • Chinese herbal medicine
  • hypnotherapy
  • a honey, beeswax and olive oil mixture
  • Nigella sativa (black seed) oil
  • gamma linolenic acid (an essential fatty acid).

No studies evaluating the effectiveness or safety of acupuncture, acupressure, meditation, relaxation techniques, naturopathy, hydrotherapy, balneology or Western herbal medicines were identified.

Homeopathy

No controlled trials evaluating the use of homeopathy in childhood atopic eczema were identified. One observational study followed children (mean age 6.7 ± 4.1 years) for a total of 24 months following an initial homeopathic consultation and course of treatment for a variety of diagnoses (n = 1130, 20% of whom had atopic eczema).469 [EL = 3] The main outcomes were child’s/parent’s and physician’s assessments (rated on a scale from 0 to 10), and quality of life at 0, 3, 12 and 24 months. All parameters improved compared with baseline at 24 months according to the child’s/parent’s and practitioner’s assessments (quality of life was assessed by parents for children under 6 years, P < 0.001). No individual data for atopic eczema were reported.

One case series reported the use of homeopathy in children and adults with predominantly mild to moderate atopic eczema (n = 36, 25% of whom were aged 11 months to 12 years).470 [EL = 3] The children received individualised homeopathic treatment between June 1995 and June 2001 in an Indian homeopathic medical college. Results were reported separately for children with skin symptoms only (n = 6), and for those with skin and respiratory symptoms (n = 3). Results were presented in terms of percentage relief/improvement. In the skin symptom only group, 3/6 were rated 99% with no new exacerbations, 2/6 were rated 60% with occasional exacerbations, and 1/6 was rated 20% (negative result) and discontinued treatment. In the skin and respiratory symptom group, 2/3 were rated 99%, 90% with no new exacerbations and 1/3 was rated 40% with new recurrence.

No safety data were identified in relation to homeopathy in children with atopic eczema.

Herbal medicine

One RCT considered the effectiveness of Chinese herbal medicine in children with atopic eczema, and a 1 year follow-up study of the same children provided longer term data.471,472 [EL = 1− and EL = 3, respectively] The RCT included 37 children with non-exudative atopic eczema with an age range of 1.5–18 years. The main outcome measures were mean severity score (0–3), erythema, surface damage and adverse events (including creatinine and endogenous steroid excretion). Median percentage changes from baseline of the clinical scores for erythema were 51% for Chinese herbs compared with 6.1% for placebo. The corresponding figures for surface damage were 63.1% and 6.2%. No safety issues were reported. The 1 year follow-up study of the children (all on active treatment) concluded that Chinese herbal medicine in the medium term proved helpful for approximately half the children who took part in the original study. However, since these studies were published, a Cochrane review has reported that the product used in the studies has ceased to be manufactured.473

A case series investigated a pentaherbs capsule treatment for atopic eczema in Chinese children (n = 9, aged 5–13.5 years).474 [EL = 3] Treatment with three pentatherb capsules was given twice daily for 4 months. After 3 months, 7/9 children had a significant reduction in their SCORAD severity score (from 60.3 to 40.0, P = 0.008). Significant differences were also noted in the extent, intensity, pruritus and sleep loss components of the SCORAD scale (P < 0.05 for all). There was no clinical or biochemical evidence of any adverse drug reaction during the study period.

A case report of a 28-year-old woman with atopic eczema who experienced two episodes of hepatitis described how the woman developed acute liver failure following the second episode and died, despite having had a liver transplant.475 The mixture she had used included two plant components (Dictamnus dasycarpus and Paeonia species) that were also contained in mixtures used by two women described in a case series. The women suffered acute hepatic illness after using traditional Chinese herbs.476 Both women recovered fully.

At least six cases of hepatotoxicity, one of cardiomyopathy and two of renal failure have been associated with the use of Chinese herbs for atopic eczema.477–479 In 1999, aristolochic acid derived from Aristolochia manshuriensis (named Mutong) was cited as the cause of renal failure in two women undertaking long-term Chinese herbal medicine for atopic eczema. Mutong is a common ingredient in Chinese therapies for atopic eczema and can also be derived from species of Akebia and Clematis which do not contain aristolochic acid. Soon after this report was published, Aristolochia species were banned in the UK.

Safety issues have also been raised concerning the adulteration of Chinese herbal treatments for childhood eczema with conventional medication.480 One letter described two case reports of children that had presented at clinic with improved atopic eczema symptoms following treatment with ‘herbal’ creams. In one case the product was found to contain 0.75 mg tablets of dexamethasone acetate and the other a potent topical corticosteroid.

Eleven Chinese herbal creams obtained from patients attending general and paediatric dermatology outpatient clinics were analysed and eight were found to contain dexamethasone at a mean concentration of 456 μg/g (range 64–1500 μg/g). All had been applied to areas of sensitive skin such as the face or flexures.481

In addition, some traditional herbal creams from Africa and Asia, such as Wau Wau cream and Abido cream, have also been found to contain potent topical steroids. Twenty-four ‘herbal’ creams submitted by 19 patients attending a paediatric dermatology clinic for atopic eczema in Birmingham (median age 3.82 years, range 0.69–7.98 years) were screened for their content.482 Reported sources of the creams included India, Pakistan, China and Tanzania either via UK-based herbalists/clinics, friends and family overseas or mail order. Seven labelled creams contained clobetasol propionate. Thirteen of 17 unnamed creams contained corticosteroids: clobetasol propionate (four), clobetasol propionate plus hydrocortisone (one), betamethasone valerate (two), clobetasone butyrate (three), hydrocortisone (one) and there was an unidentifiable corticosteroid in one. Five creams of the same brand contained approximately 20% proprietary clobetasol propionate cream in a paraffin base. In all cases, the parents were unaware that the creams contained topical corticosteroids.

In 2002, the MHRA stated that adulteration of herbal creams with corticosteroids for various skin conditions continued to be a significant problem in the UK and as a result issued a warning to the public.483

In January 2005, the MHRA reported that since 2002 the agency had investigated 17 suspected cases of illegal inclusion of corticosteroids in reportedly herbal creams, and of these seven were found to contain corticosteroids.483

Hypnotherapy

In one RCT, children with inadequately controlled atopic eczema were randomised to relaxation using hypnotherapy (focused on reducing itching), relaxation using biofeedback (no imagery included) or discussion with a psychologist (no instruction in specific techniques) for four 30 minute sessions 2, 3, 5 and 8 weeks after enrolment (n = 44, 31 analysed, age range 5–15 years).484 [EL = 1−] Four were receiving treatment with long-term oral corticosteroids. Children were stabilised on topical and oral treatments. After 20 weeks’ follow-up, changes in erythema, surface damage and lichenification were measured. Data from the two relaxation groups (hypnotherapy and biofeedback) showed a significant reduction from baseline in the severity of surface damage with time (P = 0.046) and of lichenification at 20 weeks (P = 0.02). There were no improvements over time in the discussion group.

Two case series investigated the use of hypnotherapy for atopic eczema.485,486 [EL = 3] The first involved a group of 11 children (age range 5–12 years) with established atopic eczema.485 After an initial control period, self-hypnosis was taught by a guided imagery technique with the aim of relieving itch and discomfort, and aiding relaxation. Over an 18 week period, atopic eczema was assessed by a doctor using an eczema score (maximum score 18) at six visits. The mean total eczema score decreased between most visits during the study with the median difference between visits 3 and 6 estimated to be 2.6, but this was not statistically significant (P = 0.139). In the second case series, 20 children (age range 2–15 years) with severe resistant atopic eczema were treated with hypnosis.486 Treatment consisted of an individualised tape of ‘Magic Music’ incorporating the elements of relaxation, stress management, ego strengthening, skin comfort and post-hypnotic suggestions via a 5–10 minute story metaphor with a further 5–10 minutes of music. Children and/or adults were asked to use the tapes nightly until the next clinic. Assessments of atopic eczema were made at three consecutive clinic appointments. All but one child showed immediate improvement which was maintained over the next two visits. A questionnaire was sent to the patients 18 months after receipt of the tape. Of the 12 responses to the questionnaire, ten children had maintained improvement in itching, scratching and sleep disturbance, and seven reported improvements in mood. Pictorial data only were presented in the paper. [EL = 3]

No safety data were identified for hypnotherapy.

Massage

One RCT considered massage therapy in young children with atopic eczema who were receiving standard care (mainly emollients and topical corticosteroids; n = 20, aged 2–8 years).487 [EL = 1−] A 20 minute massage with emollient was given by their parents and compared with standard care only for 1 month. Over the 1 month period, parents of massaged children reported lower anxiety levels in their children and children improved significantly on all clinical measures including erythema, scaling, lichenification, excoriation and pruritus. The control group only improved significantly on the scaling measure. No between-group analysis was undertaken.

No safety data were identified for massage therapy.

Aromatherapy

An RCT on the effect of aromatherapy in childhood atopic eczema involved 16 children who were randomised to either counselling plus massage using essential oils or counselling with massage using base oil only.488 [EL = 1−] Massage was performed by both therapist (weekly) and mothers (daily) for 8 weeks. Parents assessed daytime irritation score, night-time disturbance scores and general improvement scores. The results showed a statistically significant improvement of atopic eczema in both groups, but no intergroup differences. Post-trial continuation of aromatherapy treatments suggested that prolonged use of essential oils might cause allergic and irritant contact dermatitis.

Honey, beeswax and olive oil mixture

One controlled single-blind study evaluated a honey, beeswax and olive oil mixture for moderate to severe atopic eczema.489 [EL = 2−] The study included 21 children (aged 5–16 years) of which ten were receiving no treatment on entry to the study (group 1) and 11 were using topical beta-methasone esters (group 2). In group 1, lesions were treated with Vaseline® on the right side of the body and honey mixture on the left side. Both treatments were applied three times daily for 2 weeks. In group 2, skin lesions on the right side of the body were treated with betamethasone esters 0.1% and Vaseline (v/v 1:1) and those on the left side were treated with honey mixture and topical corticosteroid ointment (v/v 1:1). The main form of assessment was symptom scores at weeks 1 and 2 although at week 2 treatments were reassessed before continuing for a total of 6 weeks with a further reassessment of treatments at 4 weeks. In the honey mixture group, 8/10 children showed improvement after 2 weeks and 5/11 children pre-treated with betamethasone esters showed no deterioration upon a 75% reduction of topical corticosteroid doses (post-trial weeks 2–6) with honey mixture.

Nigella sativa (black seed) oil

One placebo-controlled double-blind RCT and one open-label study (reported in the same paper) investigated the effect of Nigella sativa (black seed) oil in patients with allergic diseases.490 [EL = 1− and EL = 3, respectively] The RCT involved a total of 63 patients (aged 6–17 years) of whom nine had atopic eczema.490 [EL = 1−] Treatment with black seed oil capsules (40–80 mg/kg/day) was compared with treatment with placebo oil capsules. Both treatments were taken three times daily for 8 weeks. Clinical improvement (patients’ subjective evaluation) occurred in 2/6 patients on black seed oil compared with 1/3 patients in the placebo group. No other clinical data were reported. The open-label study involved a total of 49 patients (aged 6–15 years) of whom six had atopic eczema.490 [EL = 3] All patients took two capsules of black seed oil, three times daily for 6–8 weeks. It was reported that 3/6 patients had subjective improvement of clinical symptoms, 2/6 remained unchanged and 1/6 had deterioration. Gastrointestinal adverse events were noted in 18% of participants.

Gamma linolenic acid

Four double-blind placebo-controlled RCTs investigated the effects of gamma linolenic acid on atopic eczema in children. Three of these trials involved evening primrose oil and the other involved borage oil (both sources of gamma linolenic acid). The first RCT involved children aged 2–4 years (n = 24) who received six 0.5 g evening primrose oil capsules or six 0.5 g placebo (olive oil) capsules daily for 4 weeks.491 [EL = 1+] After 4 weeks the total eczema score (incorporating signs and symptoms of eczema) improved significantly in children taking evening primrose oil (P < 0.01). Placebo-treated children’s clinical status remained largely unchanged.491

In the second RCT, children aged 7–12 years were randomised to receive evening primrose oil (six capsules of 500 mg) and fish oil (six capsules of 107 mg), or placebo (six capsules of olive oil) daily for 16 weeks (n = 62).65 [EL = 1+] Disease activity was monitored by clinical severity scores recorded by the investigator, topical corticosteroid requirement and symptom scores recorded by participants. The study also included adults, and the children’s data were not analysed separately. No improvement with active treatment was observed.65

In the third RCT, two doses of evening primrose oil (0.5 g/kg/day or 50% mix of 0.5 g/kg and placebo) were tested against placebo capsules (olive oil) in children (n = 51, mean age 4.2 years).492 [El=1+] After 8 weeks’ treatment a significant improvement in the overall severity of the clinical condition (assessed using the total eczema score) was seen in children treated with the high dose of evening primrose oil (high dose versus placebo, mean difference 0.56, P = 0.046) independently of whether the children had manifestations of IgE-mediated allergy. However, there was no significant improvement in the overall severity of the clinical condition in children treated with the low dose of evening primrose oil (low dose versus placebo, mean difference 0.51, P = 0.077). 492

None of the three RCTs of evening primrose oil reported any safety data.65,491,492

One placebo-controlled RCT investigated the effectiveness and tolerability of borage oil in children and adults with atopic eczema (n = 140)493 [EL = 1+] Sixty-nine children received two capsules twice daily (460 mg gamma linolenic acid), for 12 weeks. Data for children were not reported separately. At 12 weeks, the difference in mean improvements in SASSAD severity scores between the two groups was 1.4 (95% CI −2.2 to 5.0), indicating a non-significant benefit of placebo (P = 0.45). No significant differences were observed between treatment groups in the other assessments (symptom scores assessed on visual analogue scales, topical corticosteroid requirement, global assessment of response, adverse events and tolerability). Separate analysis of children’s and adults’ data did not indicate any difference in response. The treatments were well tolerated.

In 2002 the MHRA (then the Medicines Control Agency) withdrew the product licences (marketing authorisations) for two major evening primrose oil preparations because there was insufficient evidence for their effectiveness as medicines for treating atopic eczema.494 No concerns were expressed about safety and evening primrose oil is still available as a dietary supplement.

Cost-effectiveness

No cost-effectiveness analyses were identified, but two studies reported the costs of complementary therapies. One US study published in 1998 reported the cost of massage ($30), but did not link this with clinical outcomes.487 The other study provided an analysis of cost associated with homeopathy versus conventional therapy in Germany.495 Since this was not a UK study it is of limited relevance to the NHS setting. The cost analysis did not distinguish between children and adults or present the analysis by diagnosis. Resource use data on current health service use and use in the previous year were obtained for a subgroup of 38% of patients. Homeopathy accounted for 10% of overall costs, and the costs did not vary significantly between groups. However, the methods of analysis of the cost data were neither conventional nor fully explained.

Evidence statement for complementary therapies

Despite the popularity of complementary therapies for atopic eczema in children there was a lack of clinical effectiveness, cost-effectiveness and safety data. The few studies that were available on homeopathy, Chinese herbal medicine, massage, hypnotherapy and aromatherapy were of poor quality, and in some cases included adults as well as children. The evidence relating to gamma linolenic acid taken in the form of evening primrose oil or borage oil suggested that it was not an effective treatment for atopic eczema. There were significant safety concerns with some complementary therapies: some traditional herbal creams were found to contain topical corticosteroids and some Chinese herbal medicines were linked to renal damage and hepatotoxicity.

From evidence to recommendations

There was insufficient evidence for any of the complementary therapies described here to make recommendations for their use in clinical practice. The GDG noted some potential benefits of the therapies considered and identified a need for further research. Despite the lack of evidence, homeopathy is already available within the NHS. Treatments with Chinese herbal medicine showed positive outcomes although there were safety issues to be considered. Some traditional Chinese herbal medicines have been associated with liver damage and even death. In addition, serious adverse events have arisen as a result of adulteration, foreign language labelling and taxonomical errors of herbal mixtures. The evidence for massage was promising, with emollients being the optimal vehicle for application. Gamma linolenic acid supplementation was shown to be safe and some patients may feel it is of benefit despite the lack of clinical evidence.

Given the public’s concern about the safety of conventional treatments for atopic eczema (which may lead them to consider complementary therapies) it is important that the public understands that ‘natural’ remedies are not necessarily safe and that some complementary therapies are potentially harmful. It is also important that appropriately designed RCTs are conducted to evaluate the effectiveness, cost-effectiveness and safety of complementary therapies for the treatment of atopic eczema in children.

Recommendations for complementary therapies (including research recommendations) are presented in Section 7.11.

7.10. Behavioural therapies

Behavioural therapy is aimed at habit reversal. In atopic eczema, behavioural therapy attempts to break the itch–scratch cycle.

The HTA on atopic eczema treatments found limited data for psychological treatments.26 The studies that were included investigated behavioural management (habit reversal), relaxation and cognitive behavioural therapies and were conducted in adults.26

No published studies evaluating the effects of habit reversal in children with atopic eczema were identified.

One controlled trial investigated the effectiveness of cognitive behavioural-based stress management training for children aged 8–16 years with atopic eczema (n = 60).496 [EL = 2−] The trial evaluated a patient education programme implemented during inpatient rehabilitation in a German hospital setting. The average SCORAD index at the start of the study was 37.80 (SD 15.54). Children either took part in a multi-modal patient education programme or standard patient education training. The multi-modal programme was implemented in the setting of inpatient rehabilitation and consisted of ten 1–hour training sessions. Four sessions consisted of standard patient education and the remaining six comprised components of ‘anti-stress training’ in which cognitive behavioural techniques were used to modify the patients’ stress and disease management. The control group received standard education over six sessions. The outcome measures were the SCORAD index and the German coping questionnaire for children and adolescents (Stressverarbeitungsfragebogen, SVF-KJ) applied at baseline, 1 month and 6 months. Immediately after rehabilitation, both groups showed a significant reduction in disease severity (SCORAD index, P ≤ 0.001). At the 6 month assessment, there were only 44 datasets (experimental n = 25, control n = 19). The data suggested that the cognitive behavioural-based educational programme led to improvements in subjective health perception and ability to cope with common stressors. In contrast, the control group tended to cope less well with stress in the long-term.496

Educational interventions have also been used to bring about behavioural change through health education of parents of children with atopic eczema.497 Education for children with atopic eczema and their parents/carers is discussed in Section 8.1.

Cost-effectiveness

No studies that addressed the cost-effectiveness of behavioural therapy for children with atopic eczema were identified.

Evidence statement for behavioural therapies

There were no good-quality data regarding the effectiveness or cost-effectiveness of behavioural therapy in children with atopic eczema.

From evidence to recommendations

There was insufficient evidence of effectiveness or cost-effectiveness of behavioural therapy for the GDG to make a recommendation.

Research recommendations for behavioural therapy are presented in Section 7.11.

7.11. Recommendations for treatment

Recommendations for stepped approach to management

Healthcare professionals should use a stepped approach for managing atopic eczema in children. This means tailoring the treatment step to the severity of the atopic eczema. Emollients should form the basis of atopic eczema management and should always be used, even when the atopic eczema is clear. Management can then be stepped up or down, according to the severity of symptoms, with the addition of the other treatments listed in Table 7.4.

Healthcare professionals should offer children with atopic eczema and their parents or carers information on how to recognise flares of atopic eczema (increased dryness, itching, redness, swelling and general irritability). They should give clear instructions on how to manage flares according to the stepped-care plan, and prescribe treatments that allow children and their parents or carers to follow this plan.

Treatment for flares of atopic eczema in children should be started as soon as signs and symptoms appear and continued for approximately 48 hours after symptoms subside.

Research recommendations for stepped approach to management

How should flares of atopic eczema be defined/recognised, what pattern do they take and how useful is this to clinical practice?

Why this is important

Atopic eczema is an episodic disease punctuated by flares and remissions in most cases. It is important to be able to recognise the onset of a flare for children and their parents so that treatment can be given promptly and effectively thus improving quality of life and care. It would also aid decisions on clinical treatment strategies and provide an effective outcome measure for research purposes.

Which are the best, most cost-effective treatment strategies for managing and preventing flares in children with atopic eczema?

Why this is important

Atopic eczema is usually an episodic disease of exacerbation (flares) and remissions, except for severe cases where it may be continuous (2–6% of cases). Flares may occur as frequently as two or three times per month and have a very negative effect on quality of life. They are time-consuming and expensive to treat. There is limited evidence suggesting that strategies to prevent flares can reduce the number, frequency and severity of flares and the amount of treatment required. Identifying good strategies would improve patient care and quality of life, and free up NHS resources. Strategies that could be considered in this research include continuous versus intermittent topical treatments or combinations of products such as topical corticosteroids and topical calcineurin inhibitors.

What effect does improving the control of atopic eczema in the first year of life have on the long-term control and severity of atopic eczema and the subsequent development and severity of food allergy, asthma and allergic rhinitis?

Why this is important

Uncontrolled atopic eczema in children may progress to chronic disease involving the production of auto-immune antibodies to the skin. Early intervention to restore the defective skin barrier might alter the course of atopic eczema by preventing allergen penetration. A systematic review is needed to evaluate the available evidence on these factors. The results should feed in to the design of a large randomised controlled trial investigating the long-term effect of controlling atopic eczema in the first year of life. Early effective treatment to control atopic eczema and the development of other atopic conditions would be extremely cost-effective, have a major impact on service provision and improve the quality of life of children with atopic eczema and their parents and carers.

Recommendations for emollients

Healthcare professionals should offer children with atopic eczema a choice of unperfumed emollients to use every day for moisturising, washing and bathing. This should be suited to the child’s needs and preferences, and may include a combination of products or one product for all purposes. Leave-on emollients should be prescribed in large quantities (250–500 g weekly) and easily available to use at nursery, pre-school or school.

Healthcare professionals should inform children with atopic eczema and their parents or carers that they should use emollients in larger amounts and more often than other treatments. Emollients should be used on the whole body both when the atopic eczema is clear and while using all other treatments.

Healthcare professionals should inform children with atopic eczema and their parents or carers that they should use emollients and/or emollient wash products instead of soaps and detergent-based wash products.

Healthcare professionals should advise parents or carers of children aged under 12 months with atopic eczema to use emollients and/or emollient wash products instead of shampoos for the child. If shampoo is used for older children with atopic eczema it should be unperfumed and ideally labelled as being suitable for eczema; washing the hair in bath water should be avoided.

Healthcare professionals should show children with atopic eczema and their parents or carers how to apply emollients, including how to smooth emollients onto the skin rather than rubbing them in.

Healthcare professionals should offer an alternative emollient if a particular emollient causes irritation or is not acceptable to a child with atopic eczema.

Healthcare professionals should review repeat prescriptions of individual products and combinations of products with children with atopic eczema and their parents or carers at least once a year to ensure that therapy remains optimal.

Where emollients (excluding bath emollients) and other topical products are used at the same time of day to treat atopic eczema in children, the different products should ideally be applied one at a time with several minutes between applications where practical. The preferences of the child and parents or carers should determine which product should be applied first.

Research recommendations for emollients

Which are the most effective and cost-effective combinations of emollient products to use for the treatment of childhood atopic eczema?

Why this is important

Most children with atopic eczema have a very dry skin and early treatment with emollients makes the skin less itchy, reducing the severity of the eczema. There are numerous types and formulations of emollients but little data to suggest how they can best be used in the most effective and cost-effective way.

Does the regular use of emollients reduce the severity and frequency of flares and the need for other topical agents in the treatment of atopic eczema in children?

Why this is important

Clinical consensus suggests that this is the case but there is little good evidence for this. Confirmation would help to encourage children and their parents to comply with therapy and reduce the need for other therapies, as well as improve their quality of life.

Recommendations for topical corticosteroids

Healthcare professionals should discuss the benefits and harms of treatment with topical corticosteroids with children with atopic eczema and their parents or carers, emphasising that the benefits outweigh possible harms when they are applied correctly.

The potency of topical corticosteroids should be tailored to the severity of the child’s atopic eczema, which may vary according to body site. They should be used as follows:

  • use mild potency for mild atopic eczema
  • use moderate potency for moderate atopic eczema
  • use potent for severe atopic eczema
  • use mild potency for the face and neck, except for short-term (3–5 days) use of moderate potency for severe flares
  • use moderate or potent preparations for short periods only (7–14 days) for flares in vulnerable sites such as axillae and groin
  • do not use very potent preparations in children without specialist dermatological advice.

It is recommended that topical corticosteroids for atopic eczema should be prescribed for application only once or twice daily.*

It is recommended that where more than one alternative topical corticosteroid is considered clinically appropriate within a potency class, the drug with the lowest acquisition cost should be prescribed, taking into account pack size and frequency of application.*

Healthcare professionals should inform children with atopic eczema and their parents or carers that they should only apply topical corticosteroids to areas of active atopic eczema (or eczema that has been active within the past 48 hours), which may include areas of broken skin.

Healthcare professionals should exclude secondary bacterial or viral infection if a mild or moderately potent topical corticosteroid has not controlled the atopic eczema within 7–14 days. In children aged 12 months or over, potent topical corticosteroids should then be used for as short a time as possible and in any case for no longer than 14 days. They should not be used on the face or neck. If this treatment does not control the atopic eczema, the diagnosis should be reviewed and the child referred for specialist dermatological advice.

Potent topical corticosteroids should not be used in children aged under 12 months without specialist dermatological supervision.

Healthcare professionals who dispense topical corticosteroids should apply labels stating the potency class of the preparations to the container (for example, the tube), not the outer packaging.

Healthcare professionals should consider treating problem areas of atopic eczema with topical corticosteroids for two consecutive days per week to prevent flares, instead of treating flares as they arise, in children with frequent flares (two or three per month), once the eczema has been controlled. This strategy should be reviewed within 3–6 months to assess effectiveness.

A different topical corticosteroid of the same potency should be considered as an alternative to stepping up treatment if tachyphylaxis to a topical corticosteroid is suspected in children with atopic eczema.

*

These recommendations are taken from ‘Tacrolimus and pimecrolimus for atopic eczema’ (NICE technology appraisal guidance 82). They have been incorporated into this guideline in line with NICE procedures for developing clinical guidelines.

Research recommendations for topical corticosteroids

What are the long-term effects (when used for between 1 and 3 years) of typical use of topical corticosteroids in children with atopic eczema?

Why this is important

Around 70–80% of parents and carers of children with atopic eczema are concerned about the side effects of topical corticosteroids and this often prevents adherence to therapy (at least 25% of parents and carers report non-usage because of anxiety). Despite the fact that topical corticosteroids have been in clinical use since 1962, there are limited data on their long-term effects (greater than a few weeks) on skin thickness, hypothalamic–pituitary–adrenal (HPA) axis suppression and other side effects. Clinical consensus suggests that long-term usage, within clinically recommended dosages, appears to be safe; research confirming this would greatly improve adherence to therapy and clinical outcomes, and reduce parental anxiety. The research could include comparisons between children who use topical corticosteroids for shorter and longer periods, and with those who use other topical preparations such as emollients and topical calcineurin inhibitors.

What are the optimal treatment regimens for using topical corticosteroids in the treatment of atopic eczema in children?

Why this is important

Topical corticosteroids have been used since 1962, which predated modern randomised controlled trials (RCTs). High-quality comparative RCTs are required to provide data on the effectiveness and cost-effectiveness of various topical corticosteroids preparations in the treatment of atopic eczema in children.

Recommendations for topical calcineurin inhibitors

Topical tacrolimus and pimecrolimus are not recommended for the treatment of mild atopic eczema or as first-line treatments for atopic eczema of any severity.*

Topical tacrolimus is recommended, within its licensed indications, as an option for the second-line treatment of moderate to severe atopic eczema in adults and children aged 2 years and older that has not been controlled by topical corticosteroids, where there is a serious risk of important adverse effects from further topical corticosteroid use, particularly irreversible skin atrophy.*

Pimecrolimus is recommended, within its licensed indications, as an option for the second-line treatment of moderate atopic eczema on the face and neck in children aged 2–16 years that has not been controlled by topical corticosteroids, where there is a serious risk of important adverse effects from further topical corticosteroid use, particularly irreversible skin atrophy.*

For the purposes of this guidance, atopic eczema that has not been controlled by topical corticosteroids refers to disease that has not shown a satisfactory clinical response to adequate use of the maximum strength and potency that is appropriate for the patient’s age and the area being treated.*

It is recommended that treatment with tacrolimus or pimecrolimus be initiated only by physicians (including general practitioners) with a special interest and experience in dermatology, and only after careful discussion with the patient about the potential risks and benefits of all appropriate second-line treatment options.*

Healthcare professionals should explain to children with atopic eczema and their parents or carers that they should only apply topical calcineurin inhibitors to areas of active atopic eczema, which may include areas of broken skin.

Topical calcineurin inhibitors should not be used under occlusion (bandages and dressings) for treating atopic eczema in children without specialist dermatological advice.

For facial atopic eczema in children that requires long-term or frequent use of mild topical corticosteroids, consider stepping up treatment to topical calcineurin inhibitors.

*

These recommendations are taken from ‘Frequency of application of topical corticosteroids for atopic eczema’ (NICE technology appraisal guidance 81). They have been incorporated into this guideline in line with NICE procedures for developing clinical guidelines.

Research recommendations for topical calcineurin inhibitors

What are the most effective, cost-effective and safe ways of using combinations of topical calcineurin inhibitors with topical corticosteroids of different potencies in the treatment of atopic eczema in children, with particular reference to areas of thin skin such as the face and flexures?

Why this is important

Topical calcineurin inhibitors and topical corticosteroids are often combined in clinical practice but high-quality data are required on their safety and effectiveness/cost-effectiveness in terms of clinical benefit.

What is the effectiveness and safety of using topical calcineurin inhibitors for treating children with atopic eczema in comparison with using different potencies of topical corticosteroids and does this differ in various body sites such as the face?

Why this is important

There are few direct comparative data on the use of topical calcineurin inhibitors, particularly pimecrolimus, in different body sites and in comparison with topical corticosteroids of different potencies. Long-term use of hydrocortisone on the face is more likely to cause cutaneous atrophy than when used in other sites and topical pimecrolimus appears to be a suitable alternative. High-quality RCTs would help to answer this question.

How effective/cost-effective and safe is the use of topical tacrolimus 0.1% ointment for treating children with atopic eczema?

Why this is important

At present topical tacrolimus 0.1% ointment is not licensed for use in children under 16 years. However, clinical consensus suggests that it may be a useful, safer and probably more cost-effective alternative to, for example, long-term potent topical corticosteroids or systemic therapies for children with chronic eczema unresponsive to the 0.03% preparation of topical tacrolimus. High-quality RCTs and safety studies are required to answer this question.

What are the optimal treatment durations when using topical pimecrolimus and tacrolimus in the treatment of children with atopic eczema?

Why this is important

The topical calcineurin inhibitor formulations are new and relatively expensive with optimal treatment duration strategies not yet established. High-quality RCTs would lead to more effective/cost-effective therapy and a better use of scarce resources.

How safe are topical calcineurin inhibitors for long-term therapy (1–3 years) in the treatment of atopic eczema in children?

Why this is important

Topical calcineurin inhibitors are new drugs and safety for longer term use is not yet established. Adequately powered long-term studies in relation to tacrolimus and pimecrolimus are needed.

Recommendations for dry bandages and medicated dressings (including wet wrap therapy)

Occlusive medicated dressings and dry bandages should not be used to treat infected atopic eczema in children.

Localised medicated dressings or dry bandages can be used with emollients as a treatment for areas of chronic lichenified (localised skin thickening) atopic eczema in children.

Localised medicated dressings or dry bandages with emollients and topical corticosteroids can be used for short-term treatment of flares (7–14 days) or areas of chronic lichenified atopic eczema in children.

Whole-body (limbs and trunk) occlusive dressings (including wet wrap therapy) and whole-body dry bandages (including tubular bandages and garments) should not be used as first-line treatment for atopic eczema in children and should only be initiated by a healthcare professional trained in their use.

Whole-body (limbs and trunk) occlusive dressings (including wet wrap therapy) with topical corticosteroids should only be used to treat atopic eczema in children for 7–14 days (or for longer with specialist dermatological advice), but can be continued with emollients alone until the atopic eczema is controlled.

Research recommendations for dry bandages and medicated dressings (including wet wrap therapy)

What are the benefits and harms of the different bandaging therapies (for example, wet, dry and medicated bandages) in the treatment of atopic eczema in children?

Why this is important

Bandages are widely used to treat atopic eczema in children and many different treatment regimens are used. These treatments are expensive and time-consuming but there are few data on their clinical and cost-effectiveness and safety. Good-quality RCTs are required to evaluate benefits and harms, in particular which children benefit from such therapy and how therapies should be used.

How effective, cost-effective and safe are wet wrap dressings with emollients alone or in combination with various potencies of topical corticosteroids, for the longer term management (greater than 5 days consecutively) of atopic eczema in children and how do they compare with the use of other topical therapies alone?

Why this is important

Wet wrap dressings, usually combined with topical corticosteroid preparations, can be very effective for short-term treatment of severe eczema, but because they increase steroid absorption there is a significant risk of HPA axis suppression after 5 days’ use and an increased risk of skin infection. In clinical practice they are frequently used for periods longer than 5 days, with emollients alone or in combination with topical corticosteroids, often diluted. It is not known how safe, effective/cost-effective or practical they are for longer term management in comparison with using topical treatments alone.

How effective is the use of topical corticosteroids of different potencies or topical calcineurin inhibitors under occlusion for the treatment of atopic eczema in children and, if effective, for how long can they safely be used?

Why this is important

Occlusion increases absorption of a drug but this also increases the systemic effects. Increasing the effectiveness may compromise safety, particularly if a large surface area is involved. Such research would help to ascertain safety and efficacy of occlusion, particularly in the case of the topical calcineurin inhibitors, where there are no clinical data and little clinical experience of such use.

Recommendations for antihistamines and other antipruritics

Oral antihistamines should not be used routinely in the management of atopic eczema in children.

Healthcare professionals should offer a 1 month trial of a non-sedating antihistamine to children with severe atopic eczema or children with mild or moderate atopic eczema where there is severe itching or urticaria. Treatment can be continued, if successful, while symptoms persist, and should be reviewed every 3 months.

Healthcare professionals should offer a 7–14 day trial of an age-appropriate sedating antihistamine to children aged 6 months or over during an acute flare of atopic eczema if sleep disturbance has a significant impact on the child or parents or carers. This treatment can be repeated during subsequent flares if successful.

Research recommendations for antihistamines and other antipruritics

What is the clinical effectiveness, cost-effectiveness and safety of using sedating and non-sedating antihistamines in children with atopic eczema in terms of the outcomes itch and night-time sleep disturbance?

Why this is important

Antihistamines are frequently used to reduce itching and as night-time sedation for younger children with atopic eczema, often to allow parents some sleep. In school-age children the non-sedating antihistamines are sometimes used to reduce daytime itch. There are no data to support the use of antihistamines as an effective clinical strategy. However, lack of data does not mean lack of efficacy and some children describe them as helpful in reducing itch and improving sleep. This is a cost issue and important from clinical and patient perspectives.

Recommendations for treatments for infections associated with atopic eczema in children

Children with atopic eczema and their parents or carers should be offered information on how to recognise the symptoms and signs of bacterial infection with staphylococcus and/or streptococcus (weeping, pustules, crusts, atopic eczema failing to respond to therapy, rapidly worsening atopic eczema, fever and malaise). Healthcare professionals should provide clear information on how to access appropriate treatment when a child’s atopic eczema becomes infected.

Children with atopic eczema and their parents or carers should be informed that they should obtain new supplies of topical atopic eczema medications after treatment for infected atopic eczema because products in open containers can become contaminated with microorganisms and act as a source of infection.

Healthcare professionals should only take swabs from infected lesions of atopic eczema in children if they suspect microorganisms other than Staphylococcus aureus to be present, or if they think antibiotic resistance is relevant.

Systemic antibiotics that are active against Staphylococcus aureus and streptococcus should be used to treat widespread bacterial infections of atopic eczema in children for 1–2 weeks according to clinical response.

Flucloxacillin should be used as the first-line treatment for bacterial infections in children with atopic eczema for both Staphylococcus aureus and streptococcal infections. Erythromycin should be used in children who are allergic to flucloxacillin or in the case of flucloxacillin resistance. Clarithromycin should be used if erythromycin is not well tolerated.

The use of topical antibiotics in children with atopic eczema, including those combined with topical corticosteroids, should be reserved for cases of clinical infection in localised areas and used for no longer than 2 weeks.

Antiseptics such as triclosan or chlorhexidine should be used, at appropriate dilutions, as adjunct therapy to decrease bacterial load in children who have recurrent infected atopic eczema. Long-term use should be avoided.

Healthcare professionals should consider infection with herpes simplex (cold sore) virus if a child’s infected atopic eczema fails to respond to treatment with antibiotics and an appropriate topical corticosteroid.

If a child with atopic eczema has a lesion on the skin suspected to be herpes simplex virus, treatment with oral aciclovir should be started even if the infection is localised.

If eczema herpeticum (widespread herpes simplex virus) is suspected in a child with atopic eczema, treatment with systemic aciclovir should be started immediately and the child should be referred for same-day specialist dermatological advice. If secondary bacterial infection is also suspected, treatment with appropriate systemic antibiotics should also be started.

If eczema herpeticum involves the skin around the eyes, the child should be treated with systemic aciclovir and should be referred for same-day ophthalmological and dermatological advice.

Children with atopic eczema and their parents or carers should be offered information on how to recognise eczema herpeticum. Signs of eczema herpeticum are:

  • areas of rapidly worsening, painful eczema
  • clustered blisters consistent with early-stage cold sores
  • punched-out erosions (circular, depressed, ulcerated lesions) usually 1–3 mm that are uniform in appearance (these may coalesce to form larger areas of erosion with crusting)
  • possible fever, lethargy or distress.

Research recommendations for infections associated with atopic eczema in children

What are the prevalence and patterns of antibiotic resistance in children with atopic eczema and how clinically meaningful are these in terms of clinical management and the emergence of multiresistant bacteria?

Why this is important

Up to 80% of children with atopic eczema are known to harbour Staphylococcus aureus, although this may not be clinically apparent. There are data to show that there is an increasing resistance (up to 66% of cultures in some UK regions) to antibiotics such as fusidic acid, which is commonly used as a topical agent to treat infected eczema. It is not clear how important this is in clinical practice and what danger it poses to society as a whole. Much more information is required to determine the pattern and emergence of resistant strains and their relationship to the use of topical antibiotics.

How should bacterially infected atopic eczema in children be defined, how should it be treated and for how long? What are the indications for use of antimicrobial agents in terms of their clinical effectiveness (including palatability), cost-effectiveness and safety?

Why this is important

Bacterial colonisation of atopic eczema in children is common (up to 80% of cases) but not all will develop clinically manifest infection. However, secondary infection is a common cause of flares of eczema and is often unrecognised by healthcare professionals and parents/carers. Unnecessary use of antibiotics is expensive and potentially dangerous (in terms of systemic effects, development of allergy and emergence of multiresistant strains of microorganisms). Information from research is required to enable clear treatment plans to be made about when and for how long to use antimicrobial agents and which agents are the safest and most suitable for different ages of child.

Recommendations for phototherapy and systemic treatments

Healthcare professionals should consider phototherapy or systemic treatments for the treatment of severe atopic eczema in children when other management options have failed or are inappropriate and where there is a significant negative impact on quality of life. Treatment should be undertaken only under specialist dermatological supervision by staff who are experienced in dealing with children.

Phototherapy or systemic treatments should only be initiated in children with atopic eczema after assessment and documentation of severity of atopic eczema and quality of life.

Research recommendations for phototherapy and systemic treatments

How effective, cost-effective and safe is phototherapy in children with severe atopic eczema? How and when should it be used and should it be combined with other topical therapies?

Why this is important

Phototherapy is often used for children with severe atopic eczema but there are few studies reporting on its effectiveness, cost-effectiveness and long-term safety. High-quality RCTs are needed which should include comparisons with different types of phototherapy and in combination with different topical therapies.

How effective, cost-effective and safe are systemic treatment options in children with severe atopic eczema and how and when should they be used? For example: azathioprine, ciclosporin, methotrexate, mycophenolate mofetil, oral prednisolone and the newer biological agents.

Why this is important

Direct comparisons of the effectiveness of the systemic treatment options in children with severe atopic eczema are required, focusing on quality of life and long-term safety. All these treatment strategies are currently unlicensed for use in children under 12 years of age and should be restricted to specialist use.

Recommendations for complementary therapies

Children with atopic eczema and their parents or carers should be informed that the effectiveness and safety of complementary therapies such as homeopathy, herbal medicine, massage and food supplements for the management of atopic eczema have not yet been adequately assessed in clinical studies.

Children with atopic eczema and their parents or carers should be informed that:

  • they should be cautious with the use of herbal medicines in children and be wary of any herbal product that is not labelled in English or does not come with information about safe usage*
  • topical corticosteroids are deliberately added to some herbal products intended for use in children with atopic eczema*
  • liver toxicity has been associated with the use of some Chinese herbal medicines intended to treat atopic eczema.

Children with atopic eczema and their parents or carers should be asked to inform their health-care professionals if they are using or intend to use complementary therapies.

Children with atopic eczema and their parents or carers should be informed that if they plan to use complementary therapies, they should keep using emollients as well.

Children with atopic eczema and their parents or carers should be advised that regular massage with emollients may improve the atopic eczema.

*

See ‘Using herbal medicines: advice to consumers’. July 2006, MHRA, www​.mhra.gov.uk/home​/idcplg?IdcService=SS​_GET_PAGE&nodeId=661.

Research recommendation for complementary therapies

How effective, cost-effective and safe are complementary therapies for the management of atopic eczema in children and how do they compare with conventional Western therapies?

Why this is important

There are almost no data on the effectiveness of complementary treatment for children with atopic eczema, although there are some data to suggest that up to 60% of parents have tried these. High-quality RCTs are needed which should include comparisons with placebo controls and different forms of conventional and complementary medicine, used alone or in combination with each other. This will aid patient and physician choice and answer many unanswered questions. It has potential cost and licensing implications.

Research recommendations for behavioural therapies

Are behavioural and psychological interventions, for example habit reversal techniques, effective in the management of atopic eczema in children and would their use be feasible and cost-effective in clinical practice?

Why this is important

There are data to show that atopic eczema can have a negative psychological effect on children and their family. Adults with atopic eczema admit that they ‘habit scratch’, which perpetuates the disease and this is often true for children as well. There are also quality of life data to suggest that atopic eczema is worse than having other chronic childhood diseases. However, there are almost no data examining the effects of psychological interventions to treat these effects. Access for psychological help in the NHS is currently very limited and waiting lists are long. Such research would help to utilise scarce resources effectively and assist future service planning.

Copyright © 2007, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Cover of Atopic Eczema in Children
Atopic Eczema in Children: Management of Atopic Eczema in Children from Birth up to the Age of 12 Years.
NICE Clinical Guidelines, No. 57.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2007 Dec.

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...