Transfusion transmission of human T-lymphotropic virus types I and II: serologic and polymerase chain reaction results in recipients identified through look-back investigations

To determine the transmissibility of human T-lymphotropic virus types I and II (HTLV-I and HTLV-II) via transfusion, persons who, from 1983 to 1989, received blood components donated by persons who subsequently tested anti-HTLV-I-positive were evaluated. It was found that 16 (30%) of 54 evaluable recipients of transfused cellular components became infected with one of the HTLVs: 8 had HTLV-I and 8 had HTLV-II. Forty percent of platelet recipients and 28 percent of red cell recipients acquired infection. The rate of transmission of HTLV-I and HTLV-II was significantly correlated with storage age of red cell units prior to transfusion: 47 percent for red cells stored < or = 14 days and 0 for red cells stored > 14 days (p < 0.01). Multiple confirmatory serologic tests performed in 46 anti-HTLV-I enzyme immunoassay-negative recipients revealed that HTLV infection could not be excluded in 3 recipients of blood components from HTLV-II-infected donors. Polymerase chain reaction established HTLV-II infection in one recipient, and the other two recipients could not be classified with respect to HTLV infection status. It appears that some HTLV-II-infected transfusion recipients will not be detected by existing HTLV-I antigen-based reagents. If it is deemed necessary to initiate or continue look-back programs to detect transfusion transmission of HTLV-II infection, it is suggested that the current testing algorithm be modified in selected cases.