Transfusion of red blood cells from an HIV-RNA-positive/anti-HIV-negative donor without HIV infection in the recipient

More than 90 percent of recipients of HIV-1-containing blood components acquire HIV infection.1 We report the case of a recipient of red blood cells (RBCs), collected from an HIV-1-infected donor in the HIV-RNA positive/anti-HIV negative window phase, who did not develop HIV infection. In September 2004, HIV-1-specific antibodies(HIV1-2 Ab-Capture EIA, Ortho Diagnostic System,Raritan, NJ) and p24 antigen (Genescreen Plus HIV Ag-Ab, Bio-Rad Laboratories, Hercules, CA) were detected in a 44-year-old male blood donor who denied HIV-related risk behavior. The HIV-RNA assay (Cobas Amplicor HIV-1 Monitor Test, v. 1.5, Roche Diagnostics, Indianapolis, IA)confirmed HIV infection (56,000 copies/ml). Viral genotyping(AbbottViroseqTM HIV1 Genotyping SystemVersion 2.0, Abbott Molecular, Des Plaines, IL) showed an HIV-1 subtype B virus, not resistant to anti-retroviral medications. A routine blood donation 6 months earlier tested anti-HIV-negative. At that time, nucleic acid testing (NAT)for HIV was not performed in our Blood Service, and the components were released for transfusion. The RBCs were transfused 7 days after collection to the patient of this report. The plasma was transferred for processing(Kedrion, Lucca, Italy). Retrospectively, a sample that had been retained from this donation tested positive for HIVRNA(98 copies/mL). However, a sample retained from a blood donation on December 22, 2003, tested repeatedly HIV-RNA negative. Notice of the HIV-infected plasma unit was communicated immediately to Kedrion and to the recipient, a 54-year-old woman who had laparoscopic surgery. She has remained negative for HIV–RNA, p24Ag and anti-HIV when tested at 6, 12, and 14 months after transfusion. In vitro studies showed that the recipient’s peripheral blood mononuclear cells (PBMCs) could be infected with the donor’s HIV isolate. The recipient was genotyped by PCR as wild-type CCR5. Because exposure to HIV in the absence of seroconversion can be inferred by the detection of HIV-specific T lymphocytes,2,3 an in-depth immunologic analysis on the recipient was performed 14 months after the RBCs transfusion. Recipient’s PBMCs were stimulated with pools of antigens from env and gag regions of HIV. A higher than normal percentage of env-specific, IL-2-expressing CD4+ T lymphocytes (0.9%) was observed, whereas neither envspecific CD8+ nor gag-specific CD4+ or CD8+ T lymphocytes were detected. Naïve (CCR7+/CD45RA+), as well as central (CCR7+/CD45RA-) and effector (CCR7-/CD45RA-)memory CD4+ and CD8+ T lymphocytes, were measured. Both CD4+ and CD8+ naïve lymphocytes were reduced(11.3% vs. 23.4% and 10.1% vs. 22.2%) whereas CD8+ terminally-differentiated lymphocytes were augmented(22.7% vs. 7.0%) compared to results obtained in sex- and age-matched controls. These results strongly suggest that exposure to HIV antigens in the absence of replicating virus occurred in this patient. Thus, only actual infection with live and replicating virus would result in presentation of viral antigens in association with HLA class Imolecules, and elicitation of a CD8-mediated immune response. Additionally, responses to env, but not to gag, would be detected in the case of exposure toviral antigens in the absence of viral replication within the host cells. Finally, the observation that naïve cells were decreased in this patient is suggestive of a massive antigenic exposure, possibly HIV antigen-driven. Of importance, the donor’s HIV isolate was able to infect the recipient’s PBMCs and a CCR5D32 deletion was not present in either the donor or the recipient. Other investigators have reported cases illustrating that blood components may differ in their ability to transmit HIV infection.1,4 In one report, platelets from an HIVpositive donor transmitted infection, while RBCs from the same blood collection did not.5 There are several possible explanations for exposure to HIV without infection,including: 1) the presence of a low viral load at the time of donation; 2) a sub-infective volume of plasma in the RBCs; 3) loss of infectivity during storage of the component before transfusion; and 4) recipient’s resistance related to genetic background (other than the CCR5D32 deletion,which was not present in our patient). We believe that our results indicate that the presence of a strong cell-mediated immune response, possibly secondary to the exposure to a low amount of HIV, plays a role in resistance to HIV infection.