 |
Field evaluation of an HIV rapid test algorithm for the rapid diagnosis of HIV infection in Brazil
Presented by Cristine Ferreira, Brazil.
da Costa Ferreira Junior O.1, Ferreira C.2, Riedel M.3, Visinoni Widolin M.3, Sasazawa Ito T.3, Westman S.4, Brady W.4
1Federal University of Rio de Janeiro, Institute of Biology, Rio de Janeiro, Brazil, 2Ministry of Health of Brazil, Laboratory Unit, Brasilia, Brazil, 3Municipal Laboratory of Curitiba, Curitiba, Brazil, 4Centers for Disease Control and Prevention, Global Aids Programme, Brasilia, Brazil
Objectives: A recently finished phase I evaluation of HIV rapid tests (RT) in Brazil recommended the utilization of an algorithm with 3 different HIV RT for the rapid diagnosis of HIV infection. We now evaluate the field performance of this algorithm using whole blood from finger pricking in two settings: the voluntary and counseling testing site (VCT) and in ante-natal clinics (ANC).
Methods: This was a nominal linked study. A total of 1,191 finger pricked whole blood samples were analyzed from the VCT (n=628) and ANC (n=563) clients, between April 29th and October 28th, 2004. HIV diagnosis was made when the result of both screening assays [Determine HIV-1/2 (Abbott) and HIV Rapid Check (NDI-UFES, Brazil)] were concordant or, in the case of discordance, based on the result of the tiebreaker assay [Uni-Gold HIV (Trinity Biotech)]. Results of the RT algorithm were compared with those obtained with the current Brazilian algorithm for the diagnosis of HIV infection. All staff received a half day training session on finger pricking, RT performance and result interpretation.
Results: Overall, 26 positive and 1,151 negative samples were correctly diagnosed by the 3 RT algorithm. In only 3 instances the tiebreaker assay was used: in two cases the Determine assay was false-positive and in one case it was false-negative. In this case (false-negative), a technical mistake rather than lack of sensitivity of the Determine assay was the reason for the error. Problems found with the execution of the RT were associated with sample coagulation between the time from finger pricking to sample application to the RT device.
Conclusions: We have validated the 3 RT algorithm for the rapid diagnosis of HIV infection. The occurrence of a false-negative result due to human error makes the parallel testing a more appropriate algorithm.
Back to the session -
Back to the Programme-at-a-Glance
|
|