Objective-To review management of incidents involving exposure to blood reported to an occupational health unit. Design-Analysis of all reported incidents from january 1989 to June 1991. Setting-London teaching hospital. Subjects-447 health care workers and students. Main outcome measures-Immunisation against hepatitis B virus before exposure, proportion of known source patients tested for hepatitis B surface antigen and HIV antibodies, and reasons for not testing known source patients. Results-447 incidents were reported: 337 sharps injuries and 110 other exposures. 310 staff reporting incidents (205 (82%) nurses) were already immune to hepatitis B virus, nearly always because of immunisation. 345 source patients were identified, 77 of whom had already been tested for hepatitis B surface antigen (28 positive results) and 58 for HIV antibodies (18 positive results). Of those not previously tested, 145 of 266 were subsequently tested for hepatitis B surface antigen (two positive) and 149 of 287 for HIV antibodies (none positive). The main reasons for not testing source patients were that the incident was not considered a risk, that the patient had gone home, and that the clinical team were unwilling to ask the patient. Specific hepatitis B immunoglobulin was given to 18 staff who were not immune and was avoided in 11 cases by a negative result for the patient. Prophylactic zidovudine was discussed but not given to any staff member. Conclusions-Management of exposure to blood is improved by widespread immunisation against hepatitis B virus and by knowledge of source patients' hepatitis B virus and HIV status.
