Consensus Conference on Clinical Management of pediatric Atopic Dermatitis

Maintaining skin barrier function is an important therapeutic goal and a priority of treatment algorithms from the most prominent scientific societies worldwide [2‐6]. However, validated therapeutic regimens for acute and chronic phases of atopic dermatitis (AD) in children are still lacking. AD is known to be a chronic disease, which in most cases is mild or moderate but can be severe enough in some patients to require systemic therapy. The clinical picture is typically characterized by alternate periods of remissions and exacerbations, while it can also be persistent. With mild and moderate cases, a properly conducted topical therapeutic regimen is usually sufficient to obtain good control of the AD.

Topical corticosteroids (TCS) are still considered the mainstay of pharmacological treatment and the first choice drugs for AD therapy [4]. The use of TCS for eczema goes back about 60 years, and their effectiveness has been demonstrated in more than one hundred RCTs [39]. They are anti-inflammatory medications that can be applied to damaged skin to treat eczema in any stage of inflammation and to reduce itching [39‐41]. They act on a multitude of cells of the immune system, such as T lymphocytes, monocytes, macrophages and dendritic cells. It is hypothesized that TCS may act by interfering with the mechanisms of antigen processing, by binding to specific cell receptors, and by suppressing the release of pro-inflammatory cytokines [4, 41]. Taking into account these mechanisms of action, it is proposed that the antipruritic effect of TCS is best appreciated when the itching is secondary to inflammation [41]. The optimal dose of TCS can be standardized in finger tip units, a quantity of product that can be squeezed from a tube along the distal phalanx of the index finger of an adult hand; this amount containing approximately 0.5 g is the proper amount to be applied to a surface as large as two palms of the hands of an adult [42]. TCS should be applied after adequate cleansing of the area to be treated [43].

In order to achieve long-term therapeutic effects, important consideration should be give to:

However, we must always keep in mind some considerations:

Once the healing is achieved (regression of eczema and improvement or resolution of itching), the goal of therapy should be to extend the relapse-free time as long as possible. For this purpose, emollients can be effective, but patients or family members should be educated to reapply topical steroid immediately in case of relapse. The maximum monthly amount of medium or high potency TCS to be used to avoid local and/or systemic side effects is equal to 15 g in infants, 30 g in children, and 60–90 g in adults [3, 42]. In the last few years, another type of approach to maintanance care is being promoted, the so-called “proactive” approach. This new treatment strategy is particularly useful for optimizing clinical control in patients with frequent AD flare-ups. The “proactive” therapy should be started after eczema heals. It consists of applying TCS intermittently once or twice a week in the areas most prone to relapse, even when no inflammatory lesions are visible. This strategy reduces the number of relapses and lengthens the interval free from symptoms more effectively than the use of emollients alone [42, 48]. The rationale of proactive therapy is to obtain control of subclinical disease by using minimal amounts of TCS [46, 49]. In patients with exudative lesions and/or erosions, the application of TCS and even emollients can cause burning and may be poorly tolerated, especially in young children. In these cases the use of compresses may reduce exudation and promote re-epithelialization of lesions. The incidence of side effects from new generation TCS is very low [4]. However, the long-term use of TCS, especially if high potency, may cause local side effects. In rare cases, systemic effects may occur, more frequently in children due to the high ratio of total body surface area to body mass, which is about 2.5 to 3 times higher than for adults [50]. Normally, just 1 % of the TCS applied supplies all of its therapeutic action; the remaining 99 % is removed from the skin surface by rubbing, washing and exfoliation [50]. Despite this, prolonged exposure to high potency or older generation TCS, as may occur with occlusion, may lead to significant systemic absorption and suppression of the hypothalamic-pituitary-adrenal axis, especially if there is concurrent use of corticosteroids for other conditions (rhinitis, asthma) [51]. Main side effects of TCS are striae rubrae, skin atrophy, telangiectasia, skin burning, erythema and acneiform eruptions [52].

Among the side effects that are most feared is local skin atrophy [53]. The mechanism through which skin atrophy can be induced includes the inhibition of fibroblast proliferation, as well as reduced synthesis of collagen. Recent data are fairly reassuring about the risks of skin atrophy; in a study exploring skin thickness before and after prolonged use of moderate quantities of medium and high power topical steroids (not exceeding the above recommendations), atrophy was not observed in the treated areas [52]. Other studies have shown that skin atrophy due to TCS is a reversible phenomenon, and that healing may occur after a few weeks of TCS therapy discontinuation [54]. The possibility of contact dermatitis should be taken into consideration, especially when eczema worsens or does not respond to properly administered TCS therapy [4]; in these cases, the performance of patch tests for diagnostic purposes is indicated. Steroid-resistance may develop in case of superinfection with staphylococcus [55]. Even more rare is the observation of systemic side effects on growth, hyperglycemia, hypertension and glaucoma, reported in very old studies during systemic therapy [56] and even less frequently described for local therapies [50].

A systematic review of the literature on the side effects of TCS concluded that the safety profile of new TCS is good when used as directed [57]. However, corticophobia is commonly observed among families and even doctors [58, 59]. This is a condition to be avoided, as insufficient and inadequate applications of TCS limit ability to control dermatitis and can worsen therapeutic compliance [52, 60]. Especially in severe forms of AD, it is therefore important to implement a multidisciplinary approach, in which therapeutic education has an important role. TCSs constitute the first choice therapy for eczema, and in the majority of cases, they are able to maintain good control of the dermatitis, with benefits that greatly exceed the uncommon iatrogenic risks [50].

Skin infections in patients suffering from AD are most frequently caused by S. aureus, S. pyogenes and H. simplex virus (HSV). In patients with AD, the prevalence of colonization of the skin and/or nose by S. aureus varies from 60 to 100 %, while in control subjects without AD it varies from 5 to 30 %. A correlation between bacterial colonization and severity of eczema has also been reported in the literature [61]. Some toxins produced by S. aureus act as superantigens: they are able to produce a massive activation of T cells and contribute to exacerbation of skin lesions. The toxins also seem to induce the production of specific IgE, the activation of basophils, and consequently, the inflammatory cascade [62]. Topical antibiotic therapy is indicated for treatment of monofocal bacterial infections or confined impetigo, but not for simple colonization. Most topical antibiotics are available in two formulations: cream and ointment. Cream is preferable for treating exudative lesions, while ointment is preferable for dry lesions with a desquamative component (microbial eczema with lichenification). Fusidic acid and mupirocin are the most appropriate antibiotics; ideally they should be applied twice a day with bandage or 3 times a day without bandage for 7–10 days [63‐65]. Since mupirocin-resistant staphylococci strains have been isolated, treatment should not be prolonged over 10 days [62]. Mupirocin should not be used in children younger than 1 year, because of a lack of studies on this age group. Since 2007, Retapamulin has been approved in the US for pediatric patients above 9 months old for the treatment of impetigo with S. pyogenes and methicillin-resistant S. aureus [66]. The recommended therapeutic regimen is 2 applications a day for 5 days [67]. The effectiveness of 5 days of Retapamulin ointment therapy is comparable to that of fusidic acid used longer; however it should be reserved for strains resistant to conventional treatments. The recurrence of infections in patients with AD is frequently associated with nasal colonization by S. aureus; when nasal swab is positive for S. aureas, nasal decolonization with mupirocin (after execution of antibiogram) has been proven effective, with 2 applications a day in both nostrils for 5 days per month, for a variable period of 3–18 months [63]. The most frequently reported side effects after the application of topical antibiotics include irritation, itching, and contact dermatitis. Episodes of contact dermatitis are generally attributed to excipients or preservatives contained in the medicine (i.e. lanolin, cetyl alcohol, stearyl alcohol) [68]. The existence of multiple strains of S. aureus resistant to the most common antibiotics represents a major challenge in the treatment of staphylococcal infections. Over the past 10 years, there has been a significant increase in impetigo caused by S. aureus resistant to treatment with fusidic acid, probably due to its widespread and often inappropriate use in chronic dermatoses [68, 69]. The use of topical antibiotics in the treatment of patients with AD and impetigo permits healing of the infection with less absorption and very low risk of systemic side effects. The duration of treatment should be restricted to the treatment of impetigo to prevent any risk of sensitization and/or the development of drug resistance.

Tacrolimus (Protopic®) and pimecrolimus (Elidel®) belong to the topical immunomodulators (TIMs) class that acts by inhibiting the activity of calcineurin; they were approved in 2000 and in 2001, respectively, for the treatment of AD in adults and children older than 2 years of age [70, 71]. Tacrolimus is produced by the bacterium Streptomyces tsukubaensis, while pimecrolimus is a chemical derivative of ascomycin, produced by Streptomyces hygrospicus [72]. TIMs inhibit the activation of T cells through a highly selective mechanism of action that provides good control of the disease and a low risk of side effects. They are especially useful when long-term therapy is needed or when TCS would cause undesirable side effects (eg. on sensitive areas like the eyelids) [73]. Tacrolimus also acts on eosinophils, basophils and mast cells by blocking the production of inflammatory cytokines and decreasing the activation of T lymphocytes by the Langerhans cells, while pimecrolimus also inhibits the release of inflammatory cytokines from mast cells [72]. TIMs are indicated in children older than 2 years suffering from mild/moderate AD (pimecrolimus) or moderate/severe AD (tacrolimus) [4] and meeting one of the following requirements: a) absence of response to the first-line therapy with topical corticosteroids; b) contraindications to treatment with topical corticosteroids; c) side effects induced by the use of topical corticosteroids, such as skin atrophy or telangiectasia; d) necessity of a long-term maintenance therapy [70]. Tacrolimus is available as in ointment: in children ages 2 to 15 years the 0.03 % formulation is indicated, while in patients ages 16 years and older the 0.1 % formulation is indicated. Pimecrolimus is available as a 1 % cream and has lower percutaneous absorption than the tacrolimus ointment. TIMs should be applied twice a day for 2–3 weeks, then once a day until resolution of the skin lesions and itch symptoms. If there isn’t any improvement after 2 weeks of treatment, alternative therapeutic options must be considered. The proactive therapy illustrated for TCS can be also adopted for TIMs; the patient’s clinical response should be re-evaluated after 12 months to decide whether to continue [4, 74]. If relapse occurs, the therapy can be restarted with two applications per day. Any lymphadenopathy should be noted before starting therapy and monitored throughout therapy [75]. The use of TIMs is contraindicated in patients with primary and/or acquired immunodeficiency, suspected bacterial or viral infection, eroded and/or exuding lesions, or significant sun exposure. Pimecrolimus and tacrolimus can cause an initial and transient burning sensation and/or itching at the site of application, especially if the eczema is in its acute phase; therefore patients should always be warned about the possibility of these effects in order to increase their adherence to treatment. Initial simultaneous treatment with corticosteroids can be considered to reduce the occurrence of burning or itching. In the literature, cases of allergic contact dermatitis and granulomatous rosacea-like reactions are reported as adverse effects following the implementation of TIMs. Labial melanosis may also be noted following the use of tacrolimus 0.1 % ointment [4, 76].

Several studies showed that intermittent or continuous use of TIMs did not cause systemic immunosuppression or increase the risk of bacterial or viral infections during five years of follow up [4, 77]. In 2006, the Food and Drug Administration reported the potential carcinogenicity of TIMs after a few reports of skin cancer and lymphoma in patients treated with TIMs. However, surveillance studies conducted on the drugs showed that the number of malignancies in patients with TIMs is lower than in the general population, and a surveillance study on the pediatric population showed that the prevalence of malignancies in children treated with TIMs is comparable to that in the general pediatric population [77, 78]. Monitoring of tacrolimus and pimecrolimus blood levels is not currently recommended for the topical treatment of AD [4].

Wet dressing, or wet-wrap therapy, consists of the application of a topical medication followed by bandaging with two layers of gauze or tubular dressings, the first of which is moistened and the second of which remains dry [79]. After a brief 5-min bath with warm water, the skin is patted dry, and then the topical medication is applied. Subsequently, the first layer of gauze is moistened with warm water, excess liquid is squeezed out, the moist gauze is applied to the skin, and then a second dry bandage is applied. If possible, the moistening of the first layer can be repeated every 2–3 h throughout the day after removing the dry layer. Steam can be used to moisten the first layer, but thorough cleaning of the vaporizer is recommended. The most suitable topicals for this treatment are fluticasone propionate, methylprednisolone aceponate, mometasone furoate, hydrocortisone acetate, or prednicarbate mixed with a hydrophilic emollient to 10 % dilution (1 part of steroid and 9 parts of emollient) for the body or a 5 % dilution for face [80‐85]. Latex-free tight bandages and or elastic can be applied for 3 to 24 h at a time. Washable bandages can be used, but daily bandages are preferred. Wet-wrap dressing is a second-line treatment recommended for severe or refractory AD in patients older than 6 months of age [80, 81, 86]. No precise guidelines exist; effectiveness and the side effects are variable and depend on the age of the patient, the topical used, the time of occlusion, and the duration of treatment. The duration of treatment is variable from 2 to 14 days. The major results are obtained during the first week of treatment. Particular caution is necessary when applying TCS in patients at puberty, due to increased risk of developing striae distensae. The most dangerous side effect is systemic absorption of topical steroid, with transient increase in cortisol levels. Other possible side effects are folliculitis, bacterial or viral superinfection, and chills during the application of the wet layer if the water is not hot enough. Adherence and success can be increased through educational training [87].

Systemic corticosteroids have only a limited role in the therapeutic management of severe or difficult-to-control AD, both in children and adults [88‐92]. Despite their anti-inflammatory effect, these agents do not act directly on the recovery of the skin barrier and can cause several significant side effects [88]. Often, a rebound effect, or relapse, occurs after their suspension. Studies supporting their effectiveness are few and the study populations are small. In fact, there is only one placebo-controlled study by Schmitt et al. [89] that evaluates the effectiveness of oral prednisolone and cyclosporine vs. placebo in patients who continued their topical therapy (emollients and steroids). Only 1 patient out of 27 taking systemic corticosteroids showed persistent remission (improvement > 75 % in baseline SCORAD after 2 weeks of oral steroids and 4 weeks of follow-up). This study was stopped for the appearance of significant eczema flares in the group treated with prednisolone. A recent review of systemic treatments for AD [90] identified only two other randomized controlled trials evaluating the efficacy of systemic corticosteroids in children. The first work evaluated the effectiveness of a 4-week therapy with beclomethasone dipropionate in 26 children with severe refractory eczema and found that the average SCORAD was reduced by 22 % without serious adverse effects [91]. In the second study [93], children were treated for two weeks with oral flunisolide, and eczema severity was reduced by an average of 39 % with no relapses during the 3 weeks following discontinuation. However, the quality of both of these studies is very low, and neither studied methylprednisolone, which is the corticosteroid most widely used in clinical practice. In an open study of 7 children with very serious unresponsive eczema, methylprednisolone was administered through IV bolus at the dose of 20 mg/kg/day for 3 days [94]. Skin lesions and itching improved for a few months without notable side effects apart from a significant but transient lymphopenia. The Practical Allergy (PRACTALL) Consensus Group guidelines [95] published in 2006 suggested that patients with acute flares may benefit from a short course of systemic corticosteroids, but their long-term use, especially in children, should be avoided. Therefore, the routine use of systemic corticosteroids in children is not recommended. Short courses of therapy may be offered in special situations, such as a severe exacerbation widely affecting the body surface with intense itching, a transition period before starting systemic non-steroidal immunomodulatory drugs, or in the presence of comorbidities such as a severe asthma exacerbation [96‐98].

Alteration of the skin barrier and defects of innate immunity predispose patients with AD to the complication of infections, mainly bacterial but also fungal and viral.

Systemic immunosuppressants represent a valid therapeutic option for severe, widespread and refractory forms of AD, since systemic corticosteroids have a limited role in the long-term therapeutic management [92]. Systemic immunosuppressants should be considered if the disease is having a significant negative impact on a child’s quality of life. However, therapy with systemic immunosuppresants should be restricted to specialized centers. There are few randomized controlled trials comparing systemic therapies, so assessing the relative efficacy of each agent is difficult. Most of the literature suggests the use of cyclosporin, but azathioprine and methotrexate are also recommended.

Oral antihistamines (H1-receptor antagonists) block the effects of H1-receptor stimulation (vasodilation, erythema, edema) and have been included in the guidelines for AD treatment for a long time. However, their usefulness is controversial and debated [104, 142]. In general, the recent literature permits a short course in order to control itching. Some authors allow systemic therapy with H1 antihistamines only in patients whose AD is associated with symptoms of allergic rhino-conjunctivitis [97, 143, 144]. The pathogenesis of itch is very complex and related not only to the release of histamine [145], but also to the involvement of several other mediators such as proteases, gastrin-releasing peptide, substance P and IL-31. Experimental data show that the receptor PAR-2, present on keratinocytes and other skin cells and activated by proteases, plays a predominant role in mediating itch [146]. First-generation H1 antihistamines have lipophilic properties, which allow them to cross the blood–brain barrier and exert a subsequent sedative effect. Second-generation antihistamines have less lipophilicity and, therefore, do not cross the blood–brain barrier and are less sedating (Table 5). Due to the sedative properties of first-generation H1 antihistamines, both pediatricians and dermatologists allow short-term treatment with this medication to improve the patient’s quality of sleep [147]. On May 2014 the EMA observatory (European Medicines Agency) launched a review of drugs containing the antihistamine hydroxyzine because of possible adverse effects on the electrical conduction system of the heart [148]. In addition to H1 and H2 receptors, which are mainly located in the gastric mucosa, H3 receptors are expressed the central and (to a lesser degree) peripheral nervous system, and H4 receptors play a role in regulation of immune and inflammatory responses [149, 150]. The first results of the Early Treatment of the Atopic Child (ETAC) study [151], a multicenter trial on the efficacy of anthistamines (cetirizine/levocetirizine) to change the natural history of atopic diseases, published in 1998, showed good tolerability [152], but subsequent analysis of the benefit-risk-cost relationship was unfavorable, without no benefit in the prevention of asthma. Studies on the role of H4 receptors and their antagonists are ongoing, focusing on possible effects on the inflammatory response and, thus, on immune disorders, including AD The expression of H4 receptors on dendritic cells and keratinocytes favors a TH2-type response and the production of various mediators of inflammation, including IL-31, involved in the complex genesis of pruritus [153]. Finally, the topical use of H1 antihistamines is not recommended due to the risk of absorption and contact allergy [4].

Phototherapy is a second-line treatment, used when the AD is not controlled with emollients, topical steroids and TIMS [92, 154]. Numerous studies have demonstrated the efficacy of phototherapy in AD, both in the acute and the chronic phase [97, 155]. Currently available devices emit spectra of UVA and UVB radiation, especially narrow-band UVB (UVBTL01 311–313 nm) and UVA1 (340–400 nm). Phototherapy with UVA1 can be carried out with high (130 J cm2), medium (50 J cm2) or low -dose (10 J cm2). The mechanism of action of UV in AD is not completely understood, but an antimicrobial effect that reduces colonization of S. aureus and induces cytokine production with anti-inflammatory and immunosuppressive properties is assumed. Phototherapy with UVBTL01 is indicated in moderate forms of chronic AD, while high-dose UVA1 is preferred for severe and exuding AD. Phototherapy with UVBTL01 is most widely used due to availability, tolerability, effectiveness, and low carcinogenic risk in long-term use [155, 156]. However, it is not indicated in children younger than six years. To date, no clear guidelines for phototherapy in AD are available, and patients typically follow protocols for psoriasis (2–3 sessions per week). The starting dose is calculated based on the minimum erythema dose or Fitzpatrick skin phototype. Emollients may improve the effectiveness of treatment and topical steroids can be used to better control eczema before starting the phototherapy. Data on the duration of remission and comparing various approaches are missing. Common side effects at the beginning of the treatment include itching, burning, erythema, and folliculitis. The long-term risk of UVBTL01 phototherapy is unknown; nevertheless, its safety profile has been proven in patients with psoriasis.

Theoretically, allergen immunotherapy (AIT) may represent an additional therapeutic option for a subgroup of patients in which interventions of environmental prevention against mites (discussed below) has not led to significant improvement. The principle of AIT is administering increasing doses of clinically relevant allergen(s) in order to induce immunological tolerance in a sensitized patient [157, 158]. The therapeutic efficacy of AIT has been clearly demonstrated in allergic asthma, rhinitis and hymenoptera venom allergy [159‐162]. Despite this, AIT has not been shown to be an effective treatment for AD. In fact, while the 2012 European guidelines suggested a potential benefit of AIT in a select group of patients with coexistent allergic sensitization and AD sensitization, the latest AAD guidelines published in 2014 do not mention this therapeutic option [97].

For many years, AIT had not been considered in the treatment of AD, since it was believed to be associated with a worsening of the disease. However, the latest studies have found good tolerability of the treatment in most of patients [163‐166]. Since 1974, several clinical studies on the treatment of AD with AIT have been performed, some of which included pediatric subjects (Table 6). Otherwise, few controlled studies are available. Authors of recent reviews conclude that there is no evidence to recommend the use of AIT in the treatment of AD [167, 168]. However, some authors identify as potential candidates for AIT a subgroup of highly selected patients with the following characteristics: a) IgE-mediated sensitization (in particular to house dust mites, but also birch and grass pollens); b) severe AD (SCORAD > 50); c) Atopy Patch Test positive for the eliciting allergen. A recent meta-analysis of 8 RCTs including adult and pediatric patients found that AIT was effective, but subgroup analysis of the four studies with exclusively pediatric populations did not show significant benefits [164]. Analysis of the safety profile of the treatment found no significant difference between patients and controls, and no severe adverse reactions were reported in any of the included studies. Although the meta-analysis favored the efficacy of AIT for AD, the validity of meta-analysis’s methodology has been questioned [169]. Randomized clinical trials and observational long-term placebo-controlled studies evaluating the “disease modifying” effects of this treatment are lacking. Therefore, its therapeutic indication in patients affected exclusively by AD remains controversial [3].

The link between AD and food allergy (FA), including FA’s possible role in causing AD, remains controversial [170]. Support for the causal link comes from the observation of immediate-type allergic reactions in children with AD after the elimination and subsequent reintroduction of certain foods, especially cow’s milk (CM) and hen’s egg [171‐176]. Additionally, many studies have shown that the more severe the AD [176‐179] and earlier its first appearance [180], the greater the association with IgE-mediated FA comorbidity. The foods that most often cause FA in patients with AD are CM, egg, peanut, wheat, soy, nuts, and fish [179].

Although AD has proved to be a very complex and heterogeneous disease, it is possible to identify two main subtypes based on whether IgE levels are elevated (IgE-associated/extrinsic AD) or normal (non-IgE-associated/intrinsic AD). The latter, which is more common in preschool-age children and adults [181, 182], is associated with a lesser risk of developing asthma. According to Bergmann et al. [183], the patterns of clinical reactions to food in patients with AD can be divided into three groups: a) immediate-type reactions, usually IgE-dependent, developing within 2 h and characterized by urticaria, angioedema, rash, itching and possible involvement gastrointestinal and respiratory tract; b) delayed-type reactions that occur 6 to 48 h after the food introduction, with typical areas of eczema and are typically non-IgE mediated; c) mixed type, occuring in 40 % of children with AD and positive food challenge [184]. In 1978 a double blind controlled crossover study by Atherton et al. [185] first demonstrated an improvement in AD following a diet without CM and egg. Subsequent studies on patients with AD observed only immediate symptoms after consumption of trigger foods, suggesting a comorbid FA [172, 174, 176, 186]. Rowlands et al. [187] were able to demonstrate a link between reintroduction of previously eliminated foods and late recrudescence of AD only in one out of 17 children hospitalized with severe AD. Other studies showing the onset of late eczematous-type reactions not preceded by immediate allergic reactions are scarce and of variable quality [175, 188]. The key question is whether food allergy in the context of IgE-associated AD is an unrelated condition or whether it can trigger or worsen AD. Another option is that the food directly causes a delayed reaction. Rowlands et al. [187] assert that although it is possible that the food might induce a true eczematous lesion, this is the exception rather than the rule.

Genetic predisposition that leads to abnormalities of the skin barrier has been accompanied by an abnormal immune response that makes the skin even more vulnerable to environmental factors. Individuals with AD and concomitant filaggrin (FLG) deficiency tend to have earlier onset of the disease, which in turn is more severe and persistent and more commonly associated with allergic sensitization [189‐191]. It should however be noted that all patients with AD exhibit a barrier defect, but not all patients with FLG deficiency have AD.

In particular, a link has been observed between FLG gene mutation and peanut allergy [192]: the FLG deficiency allows food allergens to penetrate through the skin, interact with the immune system, and induce FA. In fact, although the sensitization to food classically happens through the intestines, Du Toit et al. [193] have shown that peanut sensitization can occur without previous ingestion, by simply applying peanut dissolved in oil to inflamed skin, and that early introduction decreases the frequency of peanut allergy. Also, Fox et al. [194] have found an association between environmental exposure to peanut and risk of developing peanut allergy; trace amounts of peanut can be found in furnishings or on hands [195].

The above data recently has led to the suggestion to overturn the pathogenic mechanism linking FA to AD: as opposed to FA inducing AD, exposure of skin with a barrier defect to a food may cause first a sensitization and then, later, a real FA [196‐198]. By contrast, the early consumption of food would induce, through the gastrointestinal system, a food tolerance [199‐201]. The timing and the balance of dermal and/or intestinal exposure would determine whether the child develops a food allergy or a food tolerance [202]. This mechanism has been confirmed in some mouse models, where the abrasion of the skin and subsequent deposition of peanut or ovalbumin leads to a significant specific IgE response [203, 204]; this state, in turn, can result in a clinically evident FA, as is the case with AD patients whose skin is exposed to peanut oil [193], especially if allergen doses are high. This new view of the problem could be simplistic in that it attempts to export the peanut model to other foods, in particular in CM, egg and wheat. However, it is likely that the concentration in the air of food allergens other than peanut is not as high as the concentration of peanut and inhalant allergens.

Additionally, in AD patients, the dynamics of sensitization to food allergens varies from food to food [205]; peanut acts similarly to an inhalant allergen, with a continuous increase in the concentration of specific IgE over time, while CM- and egg-specific IgE tend to decrease over time. Furthermore, the concentration of food allergens present in mattress dust varies from food to food [206]. Finally, it is known that children with AD may be sensitized to egg [207] without ever having ingested it [208]. The above is to emphasize that although the skin of the subjects with AD may allow the entry of food allergens, it is not the only point of entry for food allergens.

FA is also seen in people without AD or other skin damage [209]. In the study by Flohr et al. [210] of 619 exclusively breastfed children at 3 months of age, 154 (24.9 %) suffered from AD, but only 24 % of the babies with AD had a FLG mutation (vs. 8.4 % of those without AD). AD was found to significantly correlate with food sensitization but not with the FLG mutation [210]. Flohr et al. argue that since the children were breastfed and there are few allergens in breast milk, sensitization had occurred through the skin, not through the intestines. This argument is consistent with the previous demonstration of peanut sensitization in AD patients by application of peanut oil to the skin [193]. In our opinion, this explanation seems forced since detection of considerable quantities of heterologous proteins has been widely demonstrated in breast milk from mothers of both infants with AD as well as healthy infants [211]. It would then require individual predisposition or specific intestinal conditions to ensure that a protein first induced sensitization and then caused a real FA.

In sum, lesional skin is one of the possible routes of entry for allergens, as the direct contact of lesional skin with peanuts and/or other foods (directly spread on the skin or through contact with contaminated hands) may facilitate the occurence of sensitization. Finally, it must be considered that allergic inflammation of any type may damage the same skin barrier and inhibit the formation of FLG [212].

Several environmental factors seem to play a role in AD, particularly in exacerbations. In addition to mechanical and chemical irritants (e.g. wool, irritant soaps, toiletries containing alcohol or perfumes), indoor allergens, environmental pollutants, and climatic changes may play a role.

In patients with AD, direct contact with certain textiles with rigid fibers (e.g. wool or nylon) is a source of irritation [242], while the use of soft fabrics (e.g. knitted silk; cotton, with or without silver enrichment) may reduce the skin irritation [3, 243]. Soft fabrics reduce friction, have transpiring properties and allow the absorption of sweat and exudates, which are important in maintaining the hydrolipidic film of the skin [244]. The effectiveness of Dermasilk®, a new knitted silk fabric with antimicrobial properties, has been evaluated in a very limited number of controlled clinical trials. The studies have shown encouraging results, with improvement of clinical scores and reduction of disease recurrence during the maintenance phase [244‐247]. In a study of Gauger et al. [248], tissues enriched with silver also were effective in improving AD symptoms within 2 weeks of use. However, the risk of percutaneous absorption of silver should be always considered, especially in damaged skin [249]. ZnO-functionalized texile fibers (BenevitZink+) also seem to improve AD severity [250]. In view of the limited number of clinical studies carried out and the high cost of these materials, the most recent American guidelines conclude that there is limited evidence to support the use of these textile products in treatment of AD [97]. In any case, direct contact between irritating fabrics or wool and the skin should be avoided.

Balneotherapy is the immersion of the body or parts of it in bathtubs or pools of mineral water. Mineral waters may be categorized as sulphurous, bicarbonate, sulphate, carbonic, arsenical and ferruginous [251] and as hypotonic, isotonic or hypertonic. Temperature may vary between 20 and 40 °C. To be considered “medical,” the water must have a minimum concentration of ion and/or gas to induce relevant clinical effects. The composition and threshold dose of the thermal water may vary from one region to another and among health resorts [252]. Among the mineral waters, calcium-bicarbonate magnesium water is most indicated for AD treatment. For many years, Dead Sea waters have been used in the management of skin disease, since they are the salitiest on earth (salt content is 350 g/l vs. 40 g/l for most oceans or seas [253]) and particularly rich in salts such as MgCl2, CaCl2, KCl, and MgBr2. The presence the CaCl2 gives to the Dead Sea waters [254] a slimy feeling. The Dead Sea waters are very rich in magnesium, which promotes cell maturation and keratinocyte differentiation and is particularly useful in patients with psoriasis [255]. It is also a natural moisturizer that can improve the skin’s capability to retain water. Retrospective studies have shown the benefit of the climate and the Dead Sea salts for patients suffering from AD and other skin diseases, in the absence of significant side effects [256]. Unfortunately, few studies have been published, and they have been methodologically insufficient. The beneficial effects of treatment with Avène water for 3 weeks were evaluated in a observational study [257]; patients bathed for 20 min at 32 °C and showered to remove scales 6 days out of 7 for three consecutive weeks. At the end of the therapy, the clinical score of the AD improved significantly with long-term benefits. An open randomized controlled trial comparing the efficacy of the combination of baths of duration ≥ 4 min in a solution of 10 % Dead Sea salts plus phototherapy (exposure to 15–30 min to UVB 311 nm) vs. phototherapy alone detected greater clinical improvement in the group that bathed with the Dead Sea salts [258]. Finally, an open randomized controlled trial [259] evaluated the efficacy of the hot springs baths of Terme di Comano (Trento, Italy) in 104 children between 1 and 14 years with mild to moderate AD. Patients were alternately assigned to balneotherapy or to topical corticosteroid treatment for 2 weeks. At the end of the study, significant improvement in eczema severity and the quality of life was seen in both groups. The group treated with corticosteroids experienced more pronounced initial clinical improvement, but after 4 months, the group receiving hydrotherapy had significantly fewer and shorter exacerbations. The effects of hydrotherapy are enhanced by coexisting factors such as the relaxing spa environment [241], climatotherapy (warm weather), sun exposure (light therapy), and therapeutic education program [260]. Balneotherapy is generally not recommended during acute eczema, especially in children. Contraindications are febrile or exanthematous diseases and impetiginization. After 5–7 days of treatment, a transient exacerbation of the disease (“spa reaction”) can occur; this phenomenon is most likely to be observed if the dermatitis is not well controlled, and it usually resolves at the end of the treatment [259]. In conclusion, balneotherapy is a possible adjuvant therapy in the management of AD, but further studies are necessary to confirm its efficacy.

The complexities of AD can be summarized as follows: 1) multifactorial etiopathogenesis; 2) chronicity; 3) early age of onset; 4) involvement of the family system.

Both genetic and environmental factors contribute to its complexity. The latter are considered an expression both of the physical environment (such as health and climate) and the psycho-social system. Chronicity requires that patients (depending on age) and their caregivers adequately understand the disease and the required treatment and comply with the programmed treatment [261]. The early age of onset (0–14 years), and the varying clinical manifestations of AD at different stages of development, require physician knowledge of the psycho-social and relational implications of the disease so that interventions may be adapted accordingly. The family is a dynamic system that is critical in the emotional-affective and educational development and psychological well being of the patient. A family dealing with a chronic disease, such as AD, requires multidisciplinary support which can be summarized in the following three levels: 1) bio-pharmacological; 2) educational, pedagogical and instructive; 3) psychological/psychotherapeutical. In fact, in recent years, the scientific research on the treatment of AD has shifted from the biomedical, technical and “paternalistic” model toward a bio-psycho-social and educational one.

Therapeutic education, as defined by the WHO [262], allows us to provide not only technical information about the disease and corresponding treatments but also a customized plan developed in partnership with the patient and the patient’s caregivers.

It is important to consider how emotional factors like insecurity, inadequacy, anxiety, and depression affect atopic patients [263]. Additionally, the family is afflicted by both economic stress related to the cost of treatment and by a significant psychological burden, which is particularly heavy in young and severely affected children [264]. These factors may affect treatment compliance.

The specialist acts as the expert in therapeutic education and first conducts a psychologically supportive educational interview, covering:

Discussing challenges with treatment compliance, itching, and sleep allows for re-education about more effective strategies and improves adoption of these strategies.

A second education-oriented interview can be conducted by the broader team of physicians (dermatologist, pediatrician, allergist), nurses and psychologist with specific objectives:

These sessions can be conducted as large group demonstrations or, preferably, as smaller workshops, with more opportunity for practical individualized demonstrations [267].

After this phase, a psycho-diagnostic step may be offered for the patient and parents if the family requests it or if the staff believes it is indicated [268].

Psychotherapeutic Intervention for the patient is necessary when individual/family psychopathological suffering is severe, as may be the case in severe AD. The clinical assessment includes some psycho-diagnostic testing for parents and patients over the age of 4 years, which helps identify the most useful psychotherapeutic intervention [269]. Psychotherapeutic intervention, provided by a professional expert, supports the patient and family in coping with the emotional pain associated with the disease, improves the stability of their existential and social life, and ultimately improves adherence to treatment.

In conclusion, the paradigm of therapeutic education is the basic element of the AD Intervention Model (Fig. 1). It considers the relationship with the patient-family the essential part of the multifactorial approach, in which the clinical team works as a whole. When a team approach is not feasible, this model can still be used as a theoretical guide for the specialist, even if in a simplified form.