A prospective study on the risk of exposure to HIV during surgery in Zambia

OBJECTIVE: To investigate the relative risk of occupational HIV transmission for surgeons practising in tropical Africa compared with their western colleagues. DESIGN AND SETTING: From June to November 1993, a prospective study was performed at St Francis' Hospital, Katete, Zambia (350-bed hospital which serves a community of 300,000 people). METHODS: The HIV seroprevalence among consecutive surgical patients and the incidence of occupational parenteral exposures to blood during surgery were prospectively studied in a Zambian district hospital. HIV seroprevalence was determined by taking blood from the surgical patients on admission into the operating theatre. Serum was stored at -20 degrees C and transported to the Academic Medical Centre of the University of Amsterdam, where the presence of HIV antibodies was tested by enzyme immunoassay and seropositive samples confirmed by Western blot. Number of parenteral exposures during the study period was scored by interviewing the seven surgeons and their personnel after each surgical procedure about accidental parenteral exposures to blood. The total number of parenteral exposures per surgeon per year was obtained by extrapolation. The cumulated risk of seroconversion due to parenteral blood exposure can be calculated as: 1-(1-fp)ny, where f is the population seroprevalence, p the chance of transmission per incident (estimated to be 0.46%), n the number of parenteral exposures per year and y the years of practice. RESULTS: HIV seroprevalence in the surgical patient group was 22.3%. Twelve parenteral exposures to blood (surgeons, n = 8; other personnel, n = 4) took place in 1161 operations. Number of parenteral exposures per surgeon was extrapolated to three per year. The non-dominant index finger was exposed in 10 out of the 12 parenteral exposures. Based on these data, the risk of contracting HIV infection for a surgeon practising in Zambia for 5 years is 1.5%. The risk for a surgeon working in a western hospital when f = 0.23%, n = 20 per year (5.6% of 350 operations) and y = 5 is estimated at 0.1%. CONCLUSIONS: Although occupational exposure rate was relatively low, the HIV seroprevalence was so high that the relative cumulated seroconversion risk for surgeons in tropical Africa is estimated to be 15 times higher than in western countries. This implies that health-care organizations should bear in mind that each year one out of 300 employees working in tropical Africa may become occupationally infected with HIV.