Version May 2024
INTRODUCTION — Eosinophilic folliculitis (EF) is a pruritic skin eruption consisting of follicular papules or pustules, predominantly located on the scalp, face, neck, and upper chest (picture 1A-B).
EF is a relatively common skin eruption in patients with advanced HIV disease [1,2]. A clinically distinct form of EF (Ofuji's disease) has also been described in otherwise healthy individuals, particularly in Japan [3]. Other rare types of non-HIV-associated EF include infantile EF and EF associated with bone marrow transplantation [4,5]. Non-HIV-associated EF has also been described as a rare side effect of medication, including chemotherapy [6].
This topic reviews the presentation and management of HIV-associated EF. Bacterial folliculitis and eosinophilic pustular folliculitis of infancy are discussed separately. (See "Infectious folliculitis" and "Vesicular, pustular, and bullous lesions in the newborn and infant", section on 'Eosinophilic pustular folliculitis of infancy'.)
EPIDEMIOLOGY — HIV-associated EF most commonly occurs in patients with late-stage disease or low CD4 counts [1]. EF has been reported in adults and children infected with HIV [7-9]. The prevalence is uncertain, although one series of HIV patients reported folliculitis in 9 percent [10]. In our experience, since the advent of antiretroviral therapy, EF has become less common.
PATHOGENESIS — The etiology of EF is unknown. Clinical characteristics of EF suggest that it may be an inflammatory disease related to immune dysregulation, perhaps in association with an underlying infection [11,12].
Etiologic hypotheses include:
●Demodex mites – It has been suggested that Demodex may play a role in the pathogenesis of EF [13,14]. However, a case series of 18 patients found Demodex in only four, and in those four cases the mite was not associated with a surrounding inflammatory response, suggesting that it was not the cause of the folliculitis [15].
●Pityrosporum yeast – It has been hypothesized that persons infected with HIV who also have immunodysregulation could develop a hypersensitivity reaction to commensal follicular Pityrosporum organisms. However, histopathologic examination in 70 cases demonstrated appreciable yeast forms in just one case [15,16].
●Bacteria – Responses to antibiotics such as metronidazole have raised the possibility of a bacterial infection; however, bacteria are not consistently found in patients who respond to antibiotic treatment [17]. Additionally, most patients do not get prolonged benefits from antibiotic therapy [15].
●Autoimmunity – Based on the known association of autoimmune disease with HIV infection, and the sebaceous distribution of lesions in EF, it has been suggested that EF might be an autoimmune reaction to sebocytes or a constituent of sebum [15].
RISK FACTORS — HIV-associated EF was first described in patients with AIDS [18], and it typically occurs in late-stage disease. Patients typically have CD4 cell counts below 250 cells/mm3 [1], with a mean CD4 count of 64 cells/mm3 in a series of 18 patients [15], and a mean CD4 count of 116 cells/mm3 in a series of 57 patients [19].
CLINICAL MANIFESTATIONS — HIV-associated EF is characterized by recurrent, pruritic crops of discrete, erythematous, urticarial follicular papules and rare pustules, with a diameter of 3 to 5 mm (picture 1A-B).
The most common areas of involvement are the scalp, face, neck, and upper trunk; all are areas with a high concentration of sebaceous glands. In a series of 13 patients, all 13 had lesions on the trunk, 11 had lesions on the head and neck, and 8 had lesions on the proximal aspects of their extremities [1]. Two main patterns of distribution were identified: lesions concentrated on the trunk with a few on the face; lesions predominantly on the forehead and beard areas with limited truncal involvement. Facial involvement may be particularly common in women with EF [20].
Intense, intractable pruritus is typical [15]. True pustules are rarely observed, as the lesions are typically so pruritic that the lesions are excoriated [1,21].
Peripheral eosinophilia and elevated serum IgE are seen in about 25 to 50 percent of patients with HIV-associated EF [15,21]. Because of this low sensitivity, these tests are not routinely performed for diagnosis.
HISTOPATHOLOGY — Common histopathologic findings of EF include perifollicular infiltrates of lymphocytes and eosinophils along with spongiosis of follicular epithelium [16]. The lymphocytic and eosinophilic infiltrate is often focused at the level of the isthmus, where the sebaceous gland and duct enter the follicle [15,16]. Lysis of sebaceous ducts can be seen.
DIAGNOSIS — Although the diagnosis can be strongly suspected clinically in patients with a low CD4 count and a pruritic follicular papular eruption on the upper body, skin biopsy is needed to confirm the diagnosis [22]. Aside from HIV testing, serologic studies are not useful for diagnosis. (See "Infectious folliculitis".)
Biopsy — A 4 mm punch biopsy of an inflamed papule or pustule usually provides a sufficient specimen for evaluation. Key histopathologic findings that support EF include eosinophilic spongiosis involving the follicular epithelium and perifollicular inflammatory infiltrate containing lymphocytes and eosinophils. (See 'Histopathology' above and "Skin biopsy techniques", section on 'Punch biopsy'.)
DIFFERENTIAL DIAGNOSIS — The differential diagnosis of EF includes other follicular or pruritic disorders. The main distinction is between EF and infective folliculitis. In a prospective study of 51 HIV-positive patients with pruritic folliculitis, histopathologic examination revealed evidence for EF in 23 (45 percent), bacterial folliculitis in 21 (41 percent), Pityrosporum folliculitis in 5 (10 percent), and Demodex folliculitis in 2 (4 percent) [23]. In addition, in a series of 23 patients infected with HIV and who also had itchy folliculitis, 18 (78 percent) had EF and 5 (22 percent) had infective folliculitis [15]:
●Bacterial folliculitis – Bacterial folliculitis typically presents with erythematous papules and pustules at the sites of hair follicles (picture 2A-B). In contrast to bacterial folliculitis, intact pustules are rare in EF and EF pustules are typically smaller than pustules in bacterial folliculitis. Unlike the sterile pustules of EF, Gram stain and culture of bacterial folliculitis will demonstrate a causative organism. (See "Infectious folliculitis", section on 'Bacterial folliculitis'.)
Biopsy is not usually necessary to distinguish EF from bacterial folliculitis. However, in a series that compared 52 specimens from 50 patients with HIV-associated EF with 6 specimens of suppurative folliculitis in patients with HIV infection, biopsies of EF were distinct from those of suppurative folliculitis in that suppurative folliculitis demonstrated a predominance of neutrophils and macrophages within the infiltrate, microorganisms were easily identified, and there was often rupture of the involved follicle (picture 3A-B) [16].
●Pityrosporum folliculitis – Pityrosporum folliculitis presents with follicle-based erythematous papules and pustules distributed on the upper body, but unlike EF, rarely involves the face. A potassium hydroxide (KOH) preparation demonstrating yeast forms supports the diagnosis (picture 4). (See "Office-based dermatologic diagnostic procedures", section on 'Potassium hydroxide preparation' and "Infectious folliculitis", section on 'Fungal folliculitis'.)
Additional common disorders requiring differentiation from EF include:
●Acne vulgaris – Acne vulgaris is a common condition that, like EF, presents with papules and pustules that are primarily distributed on the upper body (picture 5A-B). Clinical features that suggest a diagnosis of acne vulgaris over EF include the presence of comedones and the absence of prominent pruritus. (See "Pathogenesis, clinical manifestations, and diagnosis of acne vulgaris".)
●Pruritic papular eruption – Pruritic papular eruption (PPE) is an intensely pruritic skin condition seen commonly in people infected with HIV in Sub-Saharan Africa. Although the etiology is uncertain, there is some evidence that it involves an exaggerated reaction to arthropod bites (picture 6) [24]. Patients present with skin-colored papules on the extremities, face, and trunk that are often excoriated [25]. Postinflammatory hyperpigmentation may also be present.
The results of a small retrospective study suggest that histologic features and immunohistochemical stains may be of value for distinguishing EF from PPE [26]. Patients with EF demonstrated more intense inflammatory infiltrates, higher counts of tissue mast cells, and higher expression levels of CD15, CD4, and CD7. Additional studies are necessary to confirm these findings.
●Scabies – Scabies can present as a widespread pruritic papular eruption (picture 7A-B). Recognition of classic areas of involvement (eg, interdigital areas), scabies burrows, or detection of the mite on a skin scraping are useful for the diagnosis of scabies. (See "Scabies: Epidemiology, clinical features, and diagnosis".)
Other pruritic or pustular disorders such as atopic dermatitis, steroid folliculitis (also known as steroid acne (picture 8A-B)), papular urticaria, and drug eruptions may also enter the differential diagnosis. (See "Insect and other arthropod bites", section on 'Papular urticaria'.)
TREATMENT — Unlike Ofuji's disease (EF in otherwise healthy individuals), HIV-associated EF does not normally respond to indomethacin [11]. Various treatments have been tried for HIV-associated EF, but there are no controlled trials comparing treatments with each other or with placebo, precluding definitive conclusions about the best approach to treatment.
First-line therapy — Antiretroviral treatment is our preferred first-line treatment for HIV-associated EF because patients in whom HIV infection responds to antiretroviral therapy generally show improvement or resolution of their EF. We also usually prescribe topical corticosteroids to aid with improvement in pruritus. We have not had much personal success with oral antihistamines for EF pruritus; however, other clinicians utilize this therapy.
Antiretroviral therapy — Although not formally studied in clinical trials, there is widespread recognition that HIV-associated EF improves or resolves in most patients treated with antiretroviral therapy (ART). (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach".)
However, there are case reports of EF flaring during the first two to six months of ART, consistent with immune reconstitution inflammatory syndrome (IRIS) [19,27]. In a series of 34 cases of EF in patients on ART, 28 (82 percent) occurred within six months of initiating ART [19] (see "Immune reconstitution inflammatory syndrome"). Patients who appear to have a flare of EF as a manifestation of IRIS are typically treated with another therapy for EF for a period of several weeks to months. (See 'Second-line therapy' below and 'Third-line therapies' below.)
Topical corticosteroids — Potent topical corticosteroids (table 1) are common first-line therapies for EF. However, data on the efficacy of topical corticosteroids are limited. Of the 38 patients with topical corticosteroid-treated HIV-associated EF identified in a review of articles published between 1965 and 2013, 47 percent had documentation of at least partial improvement with topical corticosteroid treatment [28]. In general, topical corticosteroids can decrease pruritus and improve inflammation of existing lesions but do not suppress the development of new lesions and may not fully control symptoms [11,29].
We typically instruct patients to apply a potent topical corticosteroid (eg, group 2 or group 3 (table 1)) to affected areas on the trunk and extremities twice daily for two to three weeks, after which the patient returns for reassessment. For facial involvement we prefer to use a lower-potency agent (eg, group 6 (table 1)) to reduce the risk for local corticosteroid side effects. If satisfactory improvement in pruritus is achieved, we then instruct patients to taper the topical steroid. One example of a taper is to use the agent every other day, then every third day, and so on until the topical treatment is no longer needed. If improvement is insufficient, we then add a second-line therapy. (See 'Second-line therapy' below.)
Cutaneous atrophy is a common side effect of topical corticosteroid treatment that is particularly likely to occur with the use of potent topical corticosteroids. The adverse effects of topical corticosteroids are reviewed in greater detail separately. (See "Topical corticosteroids: Use and adverse effects".)
Antihistamines — A variety of H1 and H2 blocking agents, including hydroxyzine and doxepin, are also used as first-line treatments to control pruritus. A single case report of a patient successfully treated with cetirizine led the authors to hypothesize that antieosinophilic properties possessed by cetirizine may result in increased efficacy [30]. In our experience, the efficacy of systemic antihistamines in patients with HIV-associated EF has been disappointing.
Second-line therapy — Phototherapy is a well-tolerated therapy that may improve EF in affected patients.
Phototherapy — Data from an uncontrolled study suggest that phototherapy with broadband ultraviolet B (UVB) is effective for HIV-associated EF [31-33]. In the study, which included 14 HIV-positive patients with eosinophilic folliculitis and 7 HIV-positive patients with primary pruritus, treatment of intractable pruritus with UVB phototherapy three times per week resulted in a reduction in pruritus severity scores [34]. In addition, in a series of six patients with EF who were treated with thrice-weekly broadband UVB phototherapy sessions after not responding to other therapies, all patients reported a decrease in pruritus within the first nine treatment sessions [31]. Narrowband UVB also appeared to improve HIV-associated EF in a case report; a woman who failed to respond to multiple other treatments experienced resolution of papules and pruritus with twice-weekly narrowband UVB treatments [35]. Improvement occurred within the first several treatments. However, recurrence of EF is likely after discontinuation of UVB treatment [31,35].
When treating with UVB phototherapy, we typically administer broadband or narrowband UVB two to three times per week. A response is expected within the first two to three weeks. Once sufficient improvement is achieved, we attempt to taper treatments to the lowest effective frequency. If no response is evident within two months of treatment, we discontinue UVB phototherapy and consider other treatments. (See 'Third-line therapies' below.)
Ultraviolet (UV) phototherapy is known to suppress T cell-mediated processes and also to induce activation and replication of HIV, leading some to question its safety in individuals infected with HIV [36]. However, most evidence suggests such treatment is not associated with clinical deterioration and that it can be safely used in patients infected with HIV [37,38].
Examples of other potential adverse effects are erythema, skin dryness, pruritus, and blistering. (See "UVB phototherapy (broadband and narrowband)", section on 'Short- and long-term adverse effects'.)
Although treatment with a systemic psoralen-based photosensitizer plus ultraviolet A (PUVA) has also been used successfully to treat HIV-associated EF, the increased toxicity profile of PUVA makes UVB preferable initially [11]. (See "Psoralen plus ultraviolet A (PUVA) photochemotherapy" and "Psoralen plus ultraviolet A (PUVA) photochemotherapy", section on 'Adverse effects'.)
Third-line therapies — Patients who cannot be managed effectively with the interventions above are candidates for systemic therapy. Itraconazole and oral isotretinoin are our preferred next-line therapies based upon limited data that suggest efficacy.
Itraconazole — Itraconazole may be of benefit in patients with HIV-associated EF. In an open study, 28 HIV-positive men were treated with itraconazole 100 to 400 mg daily [39]. Complete response was seen in 17 patients and partial improvement in 4 patients. All patients who improved responded within two weeks. Some patients were able to discontinue itraconazole during the three-to six-month follow-up period, while others required continued therapy with itraconazole or topical corticosteroids to maintain the response to treatment.
The effects of itraconazole on EF may be related to an anti-inflammatory effect of the drug rather than an antimycotic effect. In the study above, a patient who was switched to fluconazole had a relapse of EF, and two other patients received fluconazole without improvement, one of whom improved with itraconazole [39]. This differential response to fluconazole raises the possibility that the effect of itraconazole is due to suppression of inflammation.
Patients treated with itraconazole should have liver function tests monitored, and drug-drug interactions should be considered.
Oral isotretinoin — Data supporting the efficacy of isotretinoin in HIV-associated EF are limited to case reports and small, uncontrolled studies [23,40,41]. Isotretinoin is most commonly administered in doses of 0.5 to 1.2 mg/kg/day and typical treatment durations range from a few weeks to a few months. Responses to treatment are usually evident within the first few weeks [40].
The risk for relapse after isotretinoin therapy may be high. In a pilot study in which seven patients were treated with courses of isotretinoin ranging from 2 to 17 weeks, relapse occurred within nine months of the discontinuation of therapy in all patients, including three patients who were given multiple courses of therapy secondary to rapid initial relapses [40].
Isotretinoin is a drug that can virtually halt sebum production, and proponents of an autoimmune reaction to sebum as the etiology of HIV-associated EF suggest that this may be the explanation for its efficacy [15]. The drug is teratogenic and also can induce a variety of cutaneous and systemic adverse effects. Due to the limited data in support of the efficacy of isotretinoin, the possibility of relapse upon the cessation of therapy, and the potential adverse effects of this drug, the use of isotretinoin for HIV-associated EF should be considered carefully. In the United States, isotretinoin must be prescribed through the iPLEDGE program, an internet-based registry that was created to reduce the incidence of fetal exposure to isotretinoin. (See "Oral isotretinoin therapy for acne vulgaris".)
Other therapies — Additional topical and systemic therapies have been reported to improve EF in small numbers of patients.
Topical permethrin — Topical permethrin 5% applied daily was effective in six patients with treatment-resistant HIV-associated EF who were found to have increased numbers of Demodex mites on skin biopsy [14]. Lesions recurred when therapy was discontinued. It is uncertain whether the effect of permethrin was due to reducing numbers of Demodex mites or to other anti-inflammatory properties. We sometimes utilize permethrin in combination with systemic therapy (eg, itraconazole); however, the efficacy of such combination therapy is not established.
Topical calcineurin inhibitors — Topical tacrolimus and pimecrolimus have been reported to have some benefit in patients with Ofuji's disease [42-44]. A case series reported rapid improvement in eight patients with HIV-associated EF who were treated with once-daily topical tacrolimus 0.1%; three of the patients had developed EF in the setting of recently initiating ART [45]. The authors reported that they have not had similar success treating EF with either tacrolimus 0.03% or with pimecrolimus.
However, concerns have been raised about the safety of topical calcineurin inhibitors, with a recommendation by the United States FDA to avoid their use in patients with compromised immune systems (see "Treatment of atopic dermatitis (eczema)", section on 'Topical calcineurin inhibitors'). We suggest not treating EF with topical calcineurin inhibitors until further information on safety and efficacy is available.
Oral metronidazole — In a series of five patients with severe HIV-associated EF unresponsive to other therapies, metronidazole 250 mg three times daily for three to four weeks resulted in complete clearing [17]. Two patients had recurrences within a few months and responded to retreatment with oral metronidazole.
Oral dapsone — Dapsone has demonstrated variable success in treating Ofuji's disease and has been used in HIV-associated EF as well [11,46,47]. Dapsone suppresses the activity of myeloperoxidase within neutrophils, and there is evidence to suggest it may suppress eosinophil peroxidase within eosinophils as well [48]. The typical dose is 50 to 100 mg, once or twice a day [11].
Oral glucocorticoids — Oral glucocorticoids are occasionally used to treat HIV-associated EF [11]. Data are insufficient to determine the ideal regimen. A short treatment course beginning with 70 mg (or 1 mg/kg) of prednisone per day tapered by 5 or 10 mg per day to treatment cessation over 7 to 14 days may be effective [49].
Although oral glucocorticoids can improve HIV-associated EF, relapses are common within a few weeks after treatment is discontinued and long-term treatment is associated with risk for serious side effects. Patients with severe disease requiring continued therapy have been treated with every-other-day dosing or weekly dosing [49]. However, long-term treatment with oral glucocorticoids should be avoided whenever feasible. (See "Major adverse effects of systemic glucocorticoids".)
The decision to use systemic glucocorticoids must take into account the adverse effects of oral glucocorticoids, particularly in immunocompromised patients.
Other — A patient treated for Pneumocystis jirovecii pneumonia (PCP) with intravenous trimethoprim-sulfamethoxazole reportedly showed improvement in his EF [41]. Other treatments that have been reported to show benefit in Ofuji's disease, though not necessarily in HIV-associated EF, include oral minocycline, clofazimine, cyclosporine, and nicotine patches [11,28,50].
SUMMARY AND RECOMMENDATIONS
●Overview – HIV-associated EF is a chronic pruritic skin eruption of uncertain etiology associated with low CD4 counts and later-stage disease. (See 'Clinical manifestations' above.)
●Diagnosis – Diagnosis depends upon clinical suspicion, an appropriate presentation (intensely pruritic follicular lesions generally on the upper trunk, face, neck, or scalp), and histologic confirmation via skin biopsy of an unexcoriated lesion. (See 'Diagnosis' above.)
●First-line therapy:
•Antiretroviral therapy – We suggest use of antiretroviral therapy (ART) as the primary therapy for EF (Grade 2C). Patients who are treated with ART typically show improvement or resolution of EF. During the initial two to six months of ART, there may be a flare of EF (immune reconstitution inflammatory syndrome). (See 'Antiretroviral therapy' above.)
•Treatment of pruritus – Topical corticosteroids are useful for pruritus associated with EF. However, topical corticosteroids do not alter the course of the disease and may not fully control symptoms. Oral antihistamines are also frequently used for EF-associated pruritus. Our personal experience with antihistamines has been disappointing. (See 'Topical corticosteroids' above and 'Antihistamines' above.)
●Failure of first-line therapy – For patients who cannot be managed with ART, topical corticosteroids, or oral antihistamines, we suggest treatment with ultraviolet B (UVB) phototherapy (Grade 2C). Responses to phototherapy are usually evident within the first few weeks of treatment. Oral itraconazole and oral isotretinoin are additional treatment options for patients who fail to respond to phototherapy. (See 'Second-line therapy' above and 'Third-line therapies' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
