Risk of HIV infection from blood transfusion in Montreal

Abstract

Objectives: To determine the incidence (including associated donor characteristics and time trends) of HIV infection among repeat blood donors and to estimate the risk of HIV transmission from blood transfusion in Montreal and in Canada as a whole.

Design: Retrospective cohort analysis.

Setting: Montreal Centre Blood Transfusion Service.

Participants: People who donated blood at least twice after Nov. 1, 1985, and at least once from Apr. 1, 1989, to Mar. 31, 1993.

Intervention: Blood was screened for HIV by enzyme-linked immunosorbent assay and results were confirmed by Western blot analysis.

Outcome measures: Incidence density (the incidence rate per person-time) of HIV infection among repeat blood donors by sex, age group and region of residence, and incidence density and risk among first-time donors and for Canada as whole.

Results: There were 200,196 eligible donors and 432,631 person-years (PY) of observation. From 1989 to 1993, there were 18 HIV seroconversions among repeat donors. The crude incidence density was 3.3 per 100,000 PY (95% confidence interval [CI] 1.8 to 5.4 per 100,000 PY); it was 4.9 per 100,000 PY among men and 0.61 per 100,000 PY among women. Age-specific incidence per 100,000 PY was 2.5 among those 12-29 years of age, 5.1 among those 30-49, 2.9 among those 40-49, and 1.4 among those 50 and older. Based on an estimated mean "window period" (from when a donor's blood is capable of transmitting HIV until detectable antibody appears) of 25 days, the current risk of HIV infection from repeat donors in the window period is estimated at 1 in 440,000. Inclusion of blood units from first-time donors produces an overall risk of 1 in 390,000 (95% CI 1 in 250,000 to 655,000). The estimated risk per blood unit in Canada as a whole is 1 in 913000 (95% CI 1 in 507,000 to 2,050,000).

Conclusions: This "sentinel" population of repeat blood donors is subject to important trends in HIV spread. Therefore, estimating the incidence density of HIV infection in repeat donors provides insight into the epidemiologic characteristics of HIV infection at minimal expense. As a result of measures to improve blood safety, including HIV testing, the incidence of HIV infection among blood donors in Canada is low and the risk of HIV transmission from transfusion is extremely small, although not zero.