Folliculitis was common in the era before the introduction of highly active antiretroviral therapy (HAART), with a reported incidence of 9% at one HIV clinic [1]. In a recent study [2], folliculitis was more common in HIV-infected women with lower CD4 lymphocyte counts, and was less likely to occur in women receiving HAART independent of the CD4 cell count. Histopathological findings of folliculitis in HIV-infected individuals include acute folliculitis with bacteria or yeast, perifolliculitis with lymphocytic or mixed inflammation and eosinophilic folliculitis [3]. The best characterized form of folliculitis is HIV-associated eosinophilic folliculitis (HIV-EF), a pruritic papular eruption, involving the trunk, limbs and face. An absolute or relative peripheral eosinophilia is reported in 25–50% of affected patients [4], although this can be a manifestation of advanced HIV infection [5] The aetiology of HIV-EF is unknown. The presence of Demodex mites and Pityrosporum in some biopsies has led to suggestions that a hypersensitivity reaction to these organisms may be implicated. Other organisms and even sebum has also been proposed as causative follicular antigens [4]. Immunophenotypic analysis shows a T-cell lymphocytic infiltrate with a marked CD8 cell predominance [4]. The increased expression of the cytokines IL-4 and IL-5 in skin biopsies suggests a T helper type 2 dominant immune response [6]. Patients with HIV-EF usually have CD4 lymphocyte counts of less than 250 cells/ml [7]. There are anecdotal reports without details of the clinical improvement of HIV-EF with HAART [8,9], a decrease in the rate of HIV-EF since the introduction of HAART [10], and that HIV-EF may ‘reappear’ when the CD4 cell count approaches 200 cells/ml [8,11,12]. We describe three cases of HIV-EF in the setting of immune reconstitution after HAART (the details are summarized in Table 1). In each case antiretroviral therapy had recently been instituted, with good virological control and rising CD4 lymphocyte counts. Patients presented with pruritic lesions on the chest, neck and face. One patient had a worsening of a rash that had been present for 5 months before commencing HAART, the other two patients developed a new skin rash 4 months after commencing HAART. Skin biopsies demonstrated eosinophilic folliculitis, and in addition one biopsy revealed Demodex mites and fungal spores. The treatments used with some success included topical corticosteroids, oral tetracycline and ultraviolet therapy [7,13].Table 1: Summary of cases of HIV-associated eosinophilic folliculitis.Immune reconstitution syndrome usually occurs in the first few months after the initiation of HAART. The pathogenesis is unclear, but may involve the recovery of pathogen-specific responses resulting in the recognition of antigens associated with ongoing or past infection [14]. Bouscarat et al. [15] described ‘immune recovery inflammatory folliculitis’ in seven patients after the commencement of HAART. The lesions involved the face and upper trunk and were usually pruritic. There was no response to topical antibacterial, antifungal or antiparasitic agents. Topical corticosteroid treatment was effective. The cases were ‘acute or subacute', but the time from the commencement of therapy to the presentation of rash was not reported. The cases were associated with a mean CD4 lymphocyte count increase of 124 cells/ml. Details of skin biopsy results were not included, in particular the presence of eosinophils, but Demodex and Pityrosporum were identified in some specimens. The authors hypothesized that the immune response corresponds to the initial increase in specific memory T cells directed against pathogens (e.g. Demodex, Pityrosporum) that were already present. Delfos et al. [16] recently reported two cases of ‘Demodex folliculitis’ as a skin manifestation of immune reconstitution disease. The patients had pruritic skin eruptions on the face and trunk; one had more widespread lesions. The cases occurred 3 weeks and ‘a few days’ after commencing HAART, with increases in the CD4 lymphocyte count of 93–227 and 25–63 cells/ml, respectively. Skin scrapings revealed numerous Demodex mites. Skin biopsy results were not recorded. Topical metronidazole cream was effective in one case, and the addition of oral ivermectin was required in the other. It was suggested that uncontrolled infestation during immunodeficiency was followed by an aggravation of symptoms caused by the reconstitution of immunity and the local influx of immune-associated T cells. However, Demodex mites have been noted with various types of HIV-associated folliculitis with no ascribed causative role [3], and also without any surrounding inflammatory response [4]. We suggest that the cases of eosinophilic folliculitis described here and those noted anecdotally previously [8,11,12], as well as those reported by Bouscarat et al. [15] and Delfos et al. [16] represent a spectrum of one condition. For clarity, and to aid the future study of this phenomenon, we propose the unifying term ‘immune reconstitution folliculitis’ to apply when folliculitis occurs in the setting of immune restoration and needs to be distinguished from clinically important drug reactions or other distinctly identifiable syndromes.
Read full abstract