5.1. Background
5.1.1. Epidemiology
Seborrhoeic dermatitis is a common chronic condition that can affect people from infancy to old age; it tends to flare and remit spontaneously and is prone to recurrence after treatment. Seborrhoeic dermatitis has been reported to be associated with several conditions, including HIV (Gupta & Bluhm, 2004; Mastrolonardo et al., 2003; Maietta et al., 1990). In HIV-infected patients the prevalence is much higher and occurs early in the course of HIV disease (Wiwanitkit, 2004), with a mean CD4 count at presentation of higher than 400. The presentation can also be much more severe and/or diffuse.
5.1.2. Clinical features
Seborrhoeic dermatitis occurs in areas of the skin with a rich supply of sebaceous glands and manifests as erythematous, sharply marginated lesions with greasy scales.
Adults: Seborrhoeic dermatitis on the scalp manifests as dry, flaking desquamation (dandruff) or yellow, greasy scaling with erythema. In adolescence and adulthood it usually begins as mild greasy scaling of the scalp with erythema and scaling of the nasolabial folds or the postauricular skin. The rash morphology varies from mild scaling to an inflamed red rash. The rash often appears in areas of increased sebaceous gland activity (e.g. auricles, beard, eyebrows, trunk and flexural areas such as axilla, groin and inframammary regions). The central face may sometimes be involved. The rash rarely produces an eruption so generalized that it causes erythroderma (Janniger, 1993).
Children: In infants, seborrhoeic dermatitis may present as thick, greasy scales on the vertex of the scalp. This is commonly called cradle cap seborrhoeic dermatitis and is not included in these guidelines. As in adults, it manifests on the scalp in children as dry, flaking desquamation (dandruff) or yellow, greasy scaling with erythema. Common differential diagnoses for seborrhoeic dermatitis of the scalp are psoriasis, eczema and tinea capitis. The scales may vary in colour, appearing white, off-white or yellow. This presentation often is the only sign of seborrhoeic dermatitis in infants. However, widespread fine scaling may be seen over the scalp, face, forehead, ears and flexures and may even become generalized. Generalized seborrhoeic dermatitis is uncommon in otherwise healthy children and usually is associated with immune deficiencies. When the condition occurs in the neonatal period, it usually disappears by 6 to 12 months of age (Janniger, 1993; Janniger and Schwartz, 1995).
HIV: In HIV-infected patients, seborrhoeic dermatitis is not only more common but also more severe, sometimes with extensive spread, even covering the whole body (Chatzikokkinou et al., 2008; Marino et al., 1991). It also appears to be more refractory to conventional treatment, with more frequent relapses (Thiers, 1995). ART decreases the prevalence of seborrhoeic dermatitis (Berrey et al., 2001); in the individual patient with the condition, ART initiation may result in resolution (Dunic et al., 2004). Although ART may not cure seborrhoeic dermatitis, individuals receiving it may have less frequent and less severe occurrences (Hengge et al., 2000).
5.3. Review question and summary of evidence
The systematic review (Stephen et al., in preparation) was based on the PICO question: in children and adults living with HIV infection (receiving and not receiving ART) (P) do antifungals (ketoconazole, itraconazole, fluconazolem, bifonazole, terbinafine), calcineurin inhibitors (pimecrolimus), ART, corticosteroids or other drugs (zinc pyrithione, selinium sulphide, lithium succinate) (I) compared to each other or to no treatment (C) achieve complete resolution of the disease, or achieve remission which mainly includes symptom improvement (decrease in pruritus and rash), or a decrease in prevalence of seborrhoeic dermatitis when ART is used (O) less than or up to four weeks following commencement of treatment for short term, or greater than four weeks of treatment for medium to long term, or greater than three months of treatment for ART interventions?
Therapy for seborrhoeic dermatitis in the HIV-infected population
The number of available studies of antifungals on seborrhoeic dermatitis in the HIV-infected population is not only limited, but also the quality is poor with a serious risk of bias in almost all the studies. Only one of the nine papers identified is based on a RCT which examined the effect of lithium succinate in treatment (Langtry et al., 1997). The grade of evidence was moderate because of considerable drop out, and no evidence demonstrates the presence or absence of effect at the end of follow-up. There were no studies in children.
Impact of ART
Three prospective studies show reduced incidence of seborrhoeic dermatitis in individuals on ART. However, the studies are heterogeneous, as one study examined the effect of ART (Dunic et al., 2004), another examined recent ART (Maurer et al., 2004) and the third compared the effect of early and delayed ART (Berrey et al., 2001). Thus, the comparison groups are different. Although the quality of evidence of these studies is low, there appears to be a beneficial effect of reduced incidence of seborrhoeic dermatitis in patients on ART.
Topical therapy for seborrhoeic dermatitis in the non HIV-infected population
A systematic review by Naldi (2010) assessing topical therapies for seborrhoeic dermatitis found 12 publications (systematic reviews, RCTs or observational studies) covering topical ketoconazole 2%, bifonazole 1%, selenium sulphide 2.5%, coal tar 4% and corticosteroid clobetasol proprionate 0.05%.
All antifungal agents were more effective than the ointment they were in. The main topical antifungal agent studied was ketoconazole with moderate quality evidence of its efficacy in treatment of seborrhoeic dermatitis. The adverse effect profile was excellent, as most studies did not report any significant adverse effects with this treatment.
Although no RCTs were found for topical corticosteroids in adults with seborrhoeic dermatitis, the consensus view is that their use is effective. Current practice is to use topical corticosteroids. Affected areas of the body are treated with short courses of potent topical corticosteroids (betamethasone valerate [0.1%], mometasone furoate [0.1%]) while the face is treated with short courses of moderate potency (clobetasone butyrate [0.05%]) or low potency (hydrocortisone [1%]) corticosteroids.
Oral treatments for seborrhoeic dermatitis in the non HIV-infected population
A systematic review (Gupta et al., 2013) assessed the quantity and quality of published reports on oral therapies for seborrhoeic dermatitis. It included 21 publications (RCTs, open trials and case reports) covering itraconazole, terbinafine, fluconazole, ketoconazole, pramiconazole, prednisone, isotretinoin and homeopathic mineral therapy. For itraconazole, six open non-comparative trials and three case studies were included. All except one assessed initial treatment at 200 mg/day for seven days typically followed by varying lengths of pulse therapy for two to 11 months. The clinical improvement rate and mycological cure rate varied from 58.6% to 93.0% and 40.0% to 86% respectively. Terbinafine had higher quality trials (including two RCTs) which showed significant improvement (but not for exposed skin regions), and there was one RCT on fluconazole which showed significant effect. Although there were 21 publications for review, the quality of the evidence was low due to the absence of blinding and control groups in these studies.
5.4. Rationale for recommendations
The review by Stephen and colleagues (in preparation) focused specifically on studies in the HIV-infected population. The number of available studies of interventions for seborrhoeic dermatitis in children and adults living with HIV infection was not only limited, but also the quality was poor with a serious risk of bias in almost all of them. They therefore provided insufficient evidence on which to base a recommendation.
Because the evidence in children and adults with HIV infection was insufficient, data from studies in non HIV-infected populations were considered. The quality of this evidence was necessarily graded one level lower because of indirectness.
Although there are no studies for topical ketoconazole in seborrhoeic dermatitis specifically in HIV-infected individuals, this is the main drug that has been subject to research in the non HIV-infected population. Thise research provides strong evidence of its efficacy (Naldi, 2009; Apasrawirote et al., 2011). Although several drugs have been shown to be effective in treatment of seborrhoeic dermatitis (topical antifungal drugs and topical pimecrolimus in particular), ketoconazole is the most studied and has the strongest evidence for its effectiveness.
Almost all of the studies of antifungals (and other treatments) that provide strong evidence of their efficacy used the medication either twice or three times weekly. Although there is no clear evidence for the role of maintenance therapy in seborrhoeic dermatitis, the general consensus among the GDG was that maintenance therapy of a once-a-week application for three to four months reduced relapse rates.
Although only limited, low quality evidence is available concerning mid- to high-potency topical corticosteroids, alone or in combination with antifungals (Reygagne et al., 2007; Elewski, 2009; Shin et al,. 2009; Ortonne et al., 1992; Faergemann et al., 2007; Milani et al., 2003), there is consensus that topical corticosteroids either alone or in combination with antifungals are effective in treating seborrhoeic dermatitis of the scalp in adults. These are not recommended as first-line therapy because of low quality evidence and also because of the potential for side effects, such as striae, telangiectasia and rebound, especially with long-term use; even low-potency corticosteroids may cause adverse effects when used for long periods or on thin skin.
Data on the use of oral antifungals are limited and the evidence is very low quality even for HIV-negative individuals. No recommendation could be made because of insufficient evidence. Oral ketoconazole is no longer recommended for treatment of seborrhoeic dermatitis due to toxicity.
Only low quality evidence is available to show that ART may result in resolution of seborrhoeic dermatitis (Dunic et al., 2004). If the condition is mild and the only manifestation of HIV, the earlier initiation of lifelong ART for a condition that may respond readily to topical therapy may be less acceptable than in patients with severe and recurrent seborrhoeic dermatitis.
Although studies provided strong evidence for topical ketoconazole and moderate evidence for topical corticosteroids in seborrhoeic dermatitis, all of them were conducted in non HIV-infected populations and therefore could be considered only as indirect evidence. Therefore the quality of evidence is graded as low for topical ketoconazole and very low for topical corticosteroids.
Adverse effects
Topical ketoconazole is safe with an excellent benefit-side effect profile. Topical corticosteroids, although very effective, do have the potential for side effects, such as striae, telangiectasia and rebound, especially with long-term use. Therefore, topical ketoconazole 2% was graded as a strong recommendation for mild seborrhoeic dermatitis, and topical corticosteroids graded as a strong recommendation for severe seborrhoeic dermatitis.
Costs, availability and other implementation considerations
The data show that topical ketoconazole and topical corticosteroids are effective, easily available in all formulations (lotions, shampoos, gels, ointments, cream), affordable and acceptable to all health care providers as they have been in use for many years. Therefore, the recommendation for topical ketoconazole 2% and topical corticosteroids for the treatment of seborrhoeic dermatitis is graded as strong.
Topical ketoconazole is generally more available than other antifungals. It is easier and more feasible to use in low- and middle-income settings and less expensive compared to other treatments. Although selenium sulphide shampoo, tar shampoo, zinc pyrithione and/or mineral oil are also less expensive and easily available, all of these had lower quality evidence.
