Failure of routine HIV-1 tests in a case involving transmission with preseroconversion blood components during the infectious window period.
CONTEXT: Current screening practices for blood donations have been successful in reducing human immunodeficiency virus (HIV) transmission through receipt of contaminated blood products. However, HIV-infected blood donations made prior to seroconversion and before high levels of viral replication occur could test negative using both serologic antigen and antibody tests. Testing based on nucleic acid amplification (NAT) is being implemented to screen for HIV-infected blood donated during this period, yet the issue of single vs minipool donation screening remains unresolved. OBJECTIVES: To determine HIV-1 genetic linkage between virus in 2 HIV-1-infected recipients of blood components and virus in the donor, who was HIV antigen and antibody negative at the time of donation; to screen the blood donor's plasma with HIV NAT assays, including those currently proposed for use in US blood donation screening. DESIGN AND SETTING: Case study conducted in October 1997 involving the Communicable Disease Centre, Singapore General Hospital, and the Singapore Blood Transfusion Service, Singapore. SUBJECTS: The blood donor and the 2 recipients of donor platelets and red blood cells. MAIN OUTCOME MEASURES: Genetic analysis of the HIV-1 p17 coding region of gag and the C2V5 region of env to determine the genetic relatedness of virus from the donor and recipients; reactivity in quantitative and qualitative assays, and reactivity in donor screening HIV NAT assays in single donation and minipool screening contexts. RESULTS: Direct DNA sequencing demonstrated identical HIV-1 subtype E viral sequences in the donor and recipients. Based on comparisons of a qualitative and quantitative assay for HIV-1 RNA levels, a low level of viremia (range, 5-39 copies/mL in plasma) was estimated to be in the donor's undiluted blood at the time of donation. Additional testing using donor-screening NAT assays showed consistent detection of HIV RNA in the undiluted donor plasma whereas detection was inconsistent at the 1:16 and 1:24 dilution levels currently used in minipool screening of blood donations in the United States. CONCLUSIONS: Transmission of HIV from a blood donor to a platelet recipient and a red blood cell recipient occurred in the preseroconversion infectious window period. The viral load in the implicated donation was estimated to be less than 40 copies/mL of plasma. Current US minipool HIV NAT screening protocols may not be sufficiently sensitive to detect all infectious window-period donations. JAMA. 2000;284:210-214