A non-probabilistic selection of 100 Cuban patients at different stages of HIV infection, according to the revised classification of the Centers for Disease Control and Prevention of 1987, was made from a set of 130 persons with serologically-confirmed HIV infection. Clinical and epidemiological information about each case was collected and peripheral blood samples were taken to detect HIV-1 p24 antigen. The frequency of p24 antigenemia detection and concentration were correlated with available clinical and epidemiological data. P24 antigenemia was significantly more frequent in AIDS patients. No difference was found between the type of opportunistic diseases diagnosed in the group of patients with detectable p24 antigen and the group that was negative to antigen presence; although in the group with antigenemia concentrations over 100 pg/ml, more than one AIDS-related disease was often diagnosed simultaneously. A history of sexual intercourses with several persons who had been infected with HIV was much more frequent in patients with antigenemia, and it was associated with a shorter time elapsed from the probable date of infection to the date of their classification as AIDS patients. These results were compared with the literature review information.