Little information is available regarding the risk of human immunodeficiency virus type 1 (HIV-1) infection for patients transfused before routine anti-HIV-1 screening of blood donors was instituted in March 1985. A model was developed for estimating both the proportion and the number of transfusion recipients in the San Francisco Bay area who were infected by HIV-1 during each of the 7 years preceding routine donor screening for anti-HIV-1. The model is based on analysis of 1) donation histories of HIV-1-infected donors identified at the regional blood center; 2) HIV-1 seroprevalence estimates for homosexual and bisexual men in San Francisco; and 3) HIV-1 infection and survival rates for recipients traced by the Transfusion Safety Study and Irwin Memorial Blood Centers' Look Back Program. The incidence of transfusion-associated HIV-1 infection is estimated to have risen rapidly from the first occurrence in 1978 to a peak in late 1982 of approximately 1.1 percent per transfused unit. The decrease after 1982 coincided with the implementation of high-risk donor deferral measures. It is estimated that, overall, approximately 2135 transfusion recipients were infected with HIV-1 in the San Francisco region alone. This number suggests a higher prevalence of transfusion-associated HIV-1 infection than has been generally recognized and indicates the need for continued tracing of potentially exposed recipients. The data also strongly support the effectiveness of early donor education and self-exclusion measures and emphasize the importance of continued research and development in this area.

To evaluate the efficacy of measures for preventing the transmission of human immunodeficiency virus (HIV) by blood transfusion, trends in transfusion-associated cases of acquired immune deficiency syndrome (AIDS) reported through June 1992 were analyzed. By year of transfusion, cases rose from 56 in 1978 to 714 in 1984, dropped sharply to 288 in 1985 when screening of donated blood for HIV antibody began, and fell below 20 per year from 1986 through 1991. Reinvestigation of a sample of cases suggested that only one-fourth of those attributed in the trends analysis to post-1985 United States transfusions actually were due to that source. By year of AIDS diagnosis, cases climbed from 14 in 1982 to 824 in 1987 and subsequently remained relatively level. Of cases diagnosed in 1991 with known transfusion dates, almost all resulted from transfusions received before 1986. Cases in persons aged > or = 65 years at diagnosis fell steeply after 1987, while those in persons aged 45 to 64 years leveled and those in persons aged 25 to 44 years continued to increase; this caused the median age to decrease from 59 in 1986 to 47 in 1991. Thus, screening and other measures have almost completely prevented transmission, but, because of infections acquired before screening began, many cases continue to be diagnosed among increasingly younger persons.

No abstract