OBJECTIVES: To study the cause and clinical aspects of oral ulcers in HIV-infected patients. STUDY DESIGN: Forty-one consecutive HIV-positive patients with long-standing oral ulcers were examined; 19 were evaluated by biopsy. From these 19 cases, viral, bacterial, and fungal cultures and biopsies were taken in each patient. When indicated, special microbial stains were undertaken to identify bacteria or fungi. Ten cases without granulomatous bacterial fungal or lymphomatous features were available for in situ hybridization to detect viral DNA of herpes simplex virus 1 and 2, cytomegalovirus, varicella-zoster virus, and Epstein-Barr virus. RESULTS: Most of the oral ulcers occurred in patients with severe immunodepression. Median CD4 T-lymphocyte count was 60 cell/mm3 (range, 3 to 335). It was ascertained that nine (47%) patients had nonspecific aphthous-like ulcers, and ulcers caused by herpes group viruses were identified in six (31.5%) patients. One (5%) person was diagnosed with non-Hodgkin's lymphoma; and in one (5%) patient, multiple ulcers were an expression of lues maligna. Two ulcers (10.5%) in the palate harbored mycotic granulomatous foci (cryptococcosis, histoplasmosis). In this population, almost all of these ulcers were found to be large, persistent, and painful. CONCLUSIONS: Nontumefactive oral ulcers in HIV-positive patients may be a source of diagnostic difficulties because of the diverse array of underlying pathologic entities and multiplicity of etiologic agents. Biopsy should always be performed on long-standing ulcers because either infection or a neoplastic process may be extant. In the absence of infection or neoplasm, such lesions are then designated as ulcers not otherwise specified.
OBJECTIVES: To study the cause and clinical aspects of oral ulcers in HIV-infectedpatients. STUDY DESIGN: Forty-one consecutive HIV-positivepatients with long-standing oral ulcers were examined; 19 were evaluated by biopsy. From these 19 cases, viral, bacterial, and fungal cultures and biopsies were taken in each patient. When indicated, special microbial stains were undertaken to identify bacteria or fungi. Ten cases without granulomatous bacterial fungal or lymphomatous features were available for in situ hybridization to detect viral DNA of herpes simplex virus 1 and 2, cytomegalovirus, varicella-zoster virus, and Epstein-Barr virus. RESULTS: Most of the oral ulcers occurred in patients with severe immunodepression. Median CD4 T-lymphocyte count was 60 cell/mm3 (range, 3 to 335). It was ascertained that nine (47%) patients had nonspecific aphthous-like ulcers, and ulcers caused by herpes group viruses were identified in six (31.5%) patients. One (5%) person was diagnosed with non-Hodgkin's lymphoma; and in one (5%) patient, multiple ulcers were an expression of lues maligna. Two ulcers (10.5%) in the palate harbored mycotic granulomatous foci (cryptococcosis, histoplasmosis). In this population, almost all of these ulcers were found to be large, persistent, and painful. CONCLUSIONS: Nontumefactive oral ulcers in HIV-positivepatients may be a source of diagnostic difficulties because of the diverse array of underlying pathologic entities and multiplicity of etiologic agents. Biopsy should always be performed on long-standing ulcers because either infection or a neoplastic process may be extant. In the absence of infection or neoplasm, such lesions are then designated as ulcers not otherwise specified.