The Risk of Occupational Human Immunodeficiency Virus Infection in Health Care Workers

BACKGROUND: More than 50 cases of occupationally acquired human immunodeficiency virus (HIV) infection in health care workers (HCWs) have been reported worldwide. Determinants of injuries and of infection are important to investigate to design effective prevention programs. METHODS: In Italy, 29 acute-care public hospitals were enrolled in a multicenter study between 1986 and 1990. At each facility, all HCWs were enrolled who reported percutaneous, mucous-membrane, or nonintact-skin exposure to the body fluids and tissues to which universal precautions apply from an HIV-infected patient. Data were collected at the time of the incident on clinical status of the HIV-infected source, circumstance and type of exposure, and use of infection control precautions. The HCWs were followed up clinically and serologically for HIV infection at 1, 3, 6, and 12 months. RESULTS: A total of 1592 HIV exposures were reported in 1534 HCWs; most exposures (67%) occurred in nurses, followed by physicians and surgeons (17.5%). Needlesticks were the most common source of exposure (58.4%), followed by nonintact-skin and mucous-membrane contamination (22.7% and 11.2%, respectively) and cuts (7.7%). At the time of exposure, 77.5% of the HCWs knew or suspected that the source patient was HIV infected. Two seroconversions were observed among a total of 1488 HCWs followed up for at least 6 months: one occurred in a student nurse who had been stuck with a needle used for an HIV antibody-negative, p24 HIV antigen-positive drug addict; the other was in a nurse who experienced mucous-membrane contamination with a large quantity of blood from an HIV-positive hemophilic patient. The seroconversion rate was 0.10% after percutaneous exposure (1/1003; 95% confidence interval, 0.006% to 0.55%) and 0.63% after mucous-membrane contamination (1/158; 95% confidence interval, 0.018% to 3.47%). CONCLUSIONS: The study demonstrates a small but real risk of HIV infection after percutaneous and mucous-membrane exposure to blood of HIV-infected patients and that transmission can occur during the "window period" of infection. Furthermore, exposures to HIV are not infrequent, and many exposures could be prevented with the use of barrier precautions, appropriate behaviors, and safer devices and techniques.