BACKGROUND: Proposed testing of large plasma pools with genome amplification technology (GAT) for detection of transfusion-transmissible viruses may have unanticipated complications not associated with individual unit testing. One such potential complication, the effect of dilution resulting from pool formation, was the subject of the present study. STUDY DESIGN AND METHODS: Specimens from three plasma donor HIV type 1 (HIV-1) seroconversion panels were tested with a quantitative HIV-1 RNA GAT assay (lower detection limit, 400 copies). GAT results were compared to HIV-1 p24 antigen and anti-HIV-1/2 enzyme immunoassay results. Effects of dilution on the detection of GAT-positive panel specimens were assessed by terminal dilution with pooled volunteer-donor EDTA plasma samples. RESULTS: Low HIV-1 RNA copy numbers (755 and 890 copies/0.1-ml input) that were detectable in two individual plasma specimens before HIV-1 p24 antigen were subsequently undetectable by GAT upon dilution with an equal volume of nonreactive plasma from a single donor. HIV-1 RNA at higher copy numbers (15,500 copies/0.1-mL input) in an HIV-1 p24 antigen-reactive and anti-HIV-1/2-nonreactive specimen was undetectable when diluted to 1-in-50 (1-in-50). Terminal dilution of seven HIV-I RNA-containing plasma panel specimens indicated a proportional loss of HIV-1 RNA detectability with increasing dilution. CONCLUSION: GAT for detection of HIV-1 RNA in individual specimens was more sensitive than other HIV markers. For pooled plasma testing, GAT-independent constraints, such as effects of dilution, may preclude the use of GAT detection as a replacement for individual unit testing with HIV serologic assays.
