Epidermodysplasia verruciformis-associated human papillomaviruses and in particular human papillomavirus type 5 were recently shown to be highly prevalent in psoriatic skin. We have analyzed lesional skin from 54 psoriasis patients for infections with genital-specific and epidermodysplasia verruciformis-specific human papillomaviruses to define the spectrum of involved human papillomavirus types and to test if it is influenced by psoralen ultraviolet A therapy. Using polymerase chain reaction analysis we could detect human papillomavirus sequences in skin lesions of 83% of the tested patients. In contrast, human papillomavirus-DNA was only demonstrated in 19% of skin samples from 42 dermatologically healthy, immunocompetent individuals. Sequence analysis of the polymerase chain reaction amplimers revealed 14 human papillomavirus types, all belonging to the epidermodysplasia verruciformis or epidermodysplasia verruciformis-related papillomaviruses. Only in one case we identified sequences related to those of genital viruses, which, however, represented a putatively new human papillomavirus type. The most prevalent human papillomavirus type in our patient series was human papillomavirus type 36, found in 62% of the patients positive for human papillomavirus-DNA, followed by human papillomavirus type 5 (38%) and human papillomavirus type 38 (24%). Multiple infections with two to five different human papillomavirus types could be detected in skin samples of 63% of the analyzed patients. The overall human papillomavirus detection rate did not differ significantly between patients which have been subjected to psoralen ultraviolet A photochemotherapy or solely treated with topical preparations (77 vs 89%). Human papillomavirus type 5, however, could be detected significantly more frequent in lesions of psoralen ultraviolet A-treated patients (p < 0.001). Our data strongly argue for infections with epidermodysplasia verruciformis-specific papillomaviruses being an almost consistent feature of the lesional psoriatic skin and substantiate the importance of further studies to elucidate a possible involvement of human papillomaviruses in psoriasis pathology.