Reduced risk of transfusion-transmitted HIV in Kenya through centrally co-ordinated blood centres, stringent donor selection and effective p24 antigen-HIV antibody screening.

BACKGROUND: Following a 1994 study showing a high rate of transfusion-associated HIV, Kenya implemented WHO blood safety recommendations including: organizing the Kenya National Blood Transfusion Service (NBTS), stringent blood donor selection, and universal screening with fourth-generation p24 antigen and HIV antibody assays. Here, we estimate the risk of transfusion-associated HIV transmission in Kenya resulting from NBTS laboratory error and consider the potential safety benefit of instituting pooled nucleic acid testing (NAT) to reduce window period transmission. METHODS: From November to December 2008 in one NBTS regional centre, and from March to June 2009 in all six NBTS regional centres, every third unit of blood screened negative for HIV by the national algorithm was selected. Dried blood spots were prepared and sent to a reference laboratory for further testing, including NAT. Test results from the reference laboratory and NBTS were compared. Risk of transfusion-associated HIV transmission owing to laboratory error and the estimated yield of implementing NAT were calculated. FINDINGS: No cases of laboratory error were detected in 12,435 units tested. We estimate that during the study period, the percentage of units reactive for HIV by NAT but non-reactive by the national algorithm was 0·0% (95% exact binomial confidence interval, 0·00-0·024%). INTERPRETATION: By adopting WHO blood safety strategies for resource-limited settings, Kenya has substantially reduced the risk of transfusion-associated HIV infection. As the national testing and donor selection algorithm is effective, implementing NAT is unlikely to add a significant safety benefit. These findings should encourage other countries in the region to fully adopt the WHO strategies.

Estimated risk of HIV transmission by blood transfusion in Kenya.

BACKGROUND: During the past decade, developing countries have received limited support for blood safety programmes. The Kenya Ministry of Health did a collaborative multicentre assessment to establish the risk of HIV transmission by transfusion in Kenya, to promote awareness of blood safety issues in this country with a mature HIV epidemic, and to identify methods to reduce the risk of HIV transmission by blood transfusion in Kenya. METHODS: For 12 weeks, from April to July 1994, we collected information and blood samples from all blood donors, and pretransfusion samples were collected from all recipients in six government hospitals in Kenya. Blood donations were collected and screened for HIV according to standard practice in the hospital laboratories. Test results at a reference laboratory were compared with those of the hospital laboratories and risk of transfusion-associated HIV transmission was calculated. FINDINGS: The prevalence of HIV among blood donors was 6.4% (120 of 1877) and varied by hospital (range 2-20%). HIV test results were available for 1290 donor-recipient pairs. Of these, 26 HIV-positive donations were given to HIV-negative patients. We estimate that 2.0% of transfusions transmitted HIV. Problems in the hospitals that contributed to transfusion risk included inconsistent refrigeration, data entry errors, equipment failure, and lack of a quality-assurance programme. INTERPRETATION: A high proportion of blood transfusions transmitted HIV in this high-prevalence area of Africa, primarily because of erroneous laboratory practices. On the basis of these results, the Kenya Ministry of Health introduced a number of practical and inexpensive interventions to improve national blood safety.

Field evaluation of alternative HIV testing strategy with a rapid immunobinding assay and an agglutination assay.

A rapid immunobinding assay ('HIVCHEK', Ortho) and an agglutination assay ('Serodia-HIV', Fujirebio) were evaluated as an alternative to enzyme-linked immunosorbent assay (ELISA) and western blot under field conditions in Africa for detection of antibody to human immunodeficiency virus (HIV). 7106 specimens were tested at 25 laboratories in Kenya, Ghana, Senegal, and Zaire. HIVCHEK was used as a screening test, and serodia-HIV as a supplemental test to evaluate these assays in an alternative testing strategy to the standard ELISA/western blot testing procedure. In each country, HIVCHEK was more sensitive and specific than ELISA when compared with western blot. The sensitivity of HIVCHEK ranged from 87.0 to 96.3% and the specificity from 99.0 to 100%. The sensitivity and specificity of serodia-HIV ranged from 85 to 98% and from 88 to 98%, respectively. The sensitivity and specificity were affected by the presence of HIV-2 in Ghana and Senegal. Overall, with an HIV-1 prevalence of 14.8% in Kenya and 22.5% in Zaire, the sensitivities of the alternative strategy were 96.4% and 91.4%, the specificities 99.6% and 100%, the positive predictive values 97.6% and 100%, and the negative predictive values 99.3% and 97.9% for Kenya and Zaire, respectively. With this testing format there was an estimated average cost saving of up to 82% over the conventional strategy with ELISA/western blot. This procedure constitutes a reasonable alternative to the standard ELISA/western blot combination.