A 14-year-old boy presented with a pink firm plaque with well-defined borders in the right infra-orbital skin area. On diascopy, the infiltrate exhibited a typical apple-jelly appearance. No acid-fast bacilli could be demonstrated. A polymerase chain reaction (PCR) assay did not reveal the presence of mycobacteria in a lesional biopsy sample. Culture of biopsied tissue on Loewenstein-Jensen medium was negative. Although the tuberculosis culture and PCR did not confirm tuberculosis, a diagnosis of lupus vulgaris was made considering the clinical and histopathological findings. After a 9-month antituberculous therapy, the lesion disappeared. We believe that a diagnosis of lupus vulgaris still depends more on clinical and histopathological findings than on tuberculosis culture or PCR.