Objective: To characterize photosensitivity in HIV-infected individuals using minimal erythema dosage (MED) UVA (ultraviolet A light) and UVB (ultraviolet B light) photoprovocation light testing.
Design: Prospective, controlled analytical study.
Setting: University of California, San Francisco, between March 1995 and January 1997.
Patients: 13 HIV-seropositive patients with clinical and pathological features consistent with photodermatitis, 13 HIV-seropositive patients with biopsy-proven eosinophilic foliculitis (EF), and 10 HIV-seropositive patients with CD4 (T helper cell) count below 200 cells/uL and no history of photosensitivity or EF.
Intervention: Each patient underwent MED testing for UVB. All 13 patients with suspected photodermatitis underwent full photochallenge testing with UVA and UVB for up to 10 consecutive week days.
Results: Mean MED to UVB in patients with clinical photosensitivity and EF was lower (p = 0.004 and p = 0.022 respectively) than that of patients without a clinical history of photodermatitis. There were no significant differences in mean CD4 count or Fitzpatrick skin type. Positive photochallenge tests (papular changes at site of provocative light testing) to UVB (9 of 13 patients) were much more common than reactions to UVA (3 of 13 patients) in the photodermatitis group. All patients with clinically active photodermatitis developed papular changes at the site of UVB photochallenge testing, but only 1 of 5 patients with photodermatitis in remission developed papular changes with UVB photochallenge testing. Seven of the 13 patients with photodermatitis had Native American ancestry. Photosensitive patients were commonly taking trimethoprim-sulfamethoxazole (TMP-SMX), but no more commonly than EF or control patients.
Conclusions: Photosensitivity in HIV-infected individuals appears to be a manifestation of advanced disease. Most patients are sensitive to UVB. The most severely affected individuals are both UVB and UVA sensitive, and may show reactions to visible light. A significant Native American ancestry may be a risk factor for development of photodermatitis in patients with advanced HIV disease. Finally, patients with eosinophilic folliculitis may be subclinically photosensitive.