Prevalence of HIV-1 in blood donations following implementation of a structured blood safety policy in South Africa.

CONTEXT: The South African National Blood Service collects more than 700,000 units of blood annually from a population in which 11.4% is infected with human immunodeficiency virus 1 (HIV-1). The prevalence of HIV-1 in blood donations increased to 0.26% (1:385) in 1998, indicating that a significant number of window-period infective units were entering the blood supply (risk 3.4/100,000). OBJECTIVES: To determine whether the implementation of a new donor selection policy and educational program introduced in 1999 was associated with reductions in the incidence and prevalence of HIV-1 in blood donations and the reduced transmission risk. DESIGN: We compared the prevalence of HIV-1 in 880,534 blood donations collected from 1999 through 2000 with the 791,639 blood donations collected from 2001 through 2002. We estimated the incidence of HIV-1 in 93,378 (1999-2000) and 67,231 (2001-2002) first-time donations and the residual risk for all donations in 2001-2002 using the less-sensitive enzyme-linked immunoassay and incidence-window period model. SETTING: All blood donors in the Inland region of the South African National Blood Service were analyzed. INTERVENTION: Donor clinics in high HIV prevalence areas were closed. Programs targeting repeat donors and youth were initiated and HIV risk behavior education programs were developed. Structured donor interviews and an enhanced donor self-exclusion questionnaire were institutionalized. RESULTS: The prevalence of HIV-1 in blood donations declined from 0.17% in 1999-2000 to 0.08% in 2001-2002 after the implementation of the new donor selection and education policy. The number of high-risk donations collected decreased from 2.6% to 1.7% (P<.001), and the likelihood of these donations being infected decreased from 4.8% to 3.25%. The likelihood of first-time donors being recently infected with HIV-1 decreased from 18% to 14% (P = .07) and respective incidence of high-risk donations collected decreased from 2.6% to 1.7%. Donations from the majority black population declined from 6.6% to 4.2% (P<.001). Analysis of HIV-1 incidence in 2001-2002 suggests a residual risk of collecting a window period infectious unit of 2.6/100,000. CONCLUSION: The implementation of enhanced education and selection policies in South Africa was associated with decreased prevalence of HIV-1 in blood donations.

Relative sensitivities of licensed nucleic acid amplification tests for detection of viremia in early human immunodeficiency virus and hepatitis C virus infection.

BACKGROUND: Screening donors for human immunodeficiency virus (HIV) and hepatitis C virus (HCV) RNA is primarily performed on minipools (MPs) with one of two commercial nucleic acid amplification tests (NAT; Roche Molecular Systems; or Gen-Probe/Chiron). We compared these assays with respect to detection of RNA in early HIV and HCV infection. STUDY DESIGN AND METHODS: Twelve HIV plasma donor panels (116 serial samples) and 12 HCV panels (180 serial samples) were selected to optimally represent early viremia. Initial testing was performed in singlicate or triplicate on separately coded aliquots, both neat and at dilutions corresponding to MP screening (1:16 for Gen-Probe; 1:24 for Roche); 20 additional replicates were performed when discordant results were observed. Odds ratios (ORs) comparing detection of RNA by different assays were derived with logistic regression models. Differences in window-period closure and yields of assays in MP or individual-donation (ID) format were estimated. RESULTS: Differences in detection rates between Roche and Gen-Probe NAT assays were small and only observed with samples with very-low-level viremia. ORs for detecting RNA by the Gen-Probe versus the Roche assay were significant for HIV if conducted on MPs (1.8; 95% confidence interval [CI], 1.3-2.5) but not neat (1.0; 95% CI, 0.72-1.4). Odds of detecting HCV RNA were higher if the Gen-Probe assay was conducted either neat (2.3; 95% CI, 1.6-3.2) or on MPs (4.0; 95% CI, 2.8-5.8). These differences translated to <1 day window-period closure and

Prevalence, incidence, and residual risk of human immunodeficiency virus among community and replacement first-time blood donors in São Paulo, Brazil.

BACKGROUND: Concerted efforts have been directed toward recruitment of community rather than replacement donors in Brazil. Time trends and demographic correlates of human immunodeficiency (HIV) prevalence and incidence among first-time (FT) donors in Brazil were examined by donation type. HIV residual risk from FT-donor transfusions, and projected yield of p24 antigen and nucleic acid test (NAT) screening were estimated. STUDY DESIGN AND METHODS: HIV prevalence data and seroreactive specimens were obtained at Fundação Pró-Sangue/Hemocentro-de-São Paulo from 1995 to 2001. To estimate incidence, confirmed-positive samples from July 1998 through December 2001 were tested with a less-sensitive (detuned) enzyme immunoassay to detect recent seroconversions. Incidence data were used to estimate residual risk and p24 and NAT yield based on published window periods (WPs). RESULTS: HIV prevalence was 22 percent higher among the FT community donors than replacement donors (19.6 vs. 16.1 per 10,000; p < 0.01) and 48 percent higher among men than women (19.1 vs. 12.9; p < 0.01). In the multivariable logistic regression, both variables remained significant predictors of HIV prevalence. HIV prevalence decreased from 20.4 (1995) to 13.1 per 10,000 FT donations (2001). HIV incidence was 2.7 per 10,000 person-years. The estimated rate of infected antibody-negative donations was 14.9 per 1,000,000 units (95% confidence interval, 9.8-20.0). It was estimated that addition of p24 antigen, minipool NAT, and individual-donation NAT assays would detect 3.9 (2.0-5.8), 8.3 (5.3-11.3), and 10.8 (7.1-14.5) WP units per 1,000,000 FT donations, respectively. CONCLUSION: HIV incidence and residual transfusion risk estimates are approximately 10 times higher in Brazil FT donors compared to US and European FT donors. Community FT donors had higher HIV prevalence than replacement FT donors. The yield of p24 antigen or RNA screening will be low in Brazilian donors, but substantially higher than in US donors.

Hepatitis B and C viral load changes following initiation of highly active antiretroviral therapy (HAART) in patients with advanced HIV infection.

Chronic infection with either hepatitis B (HBV) or hepatitis C virus (HCV) is frequently present in patients seropositive for human immunodeficiency virus (HIV) because of shared routes of transmission. With the advent of highly active antiretroviral therapy (HAART) regimens capable of controlling HIV replication and dramatically prolonging the survival of HIV-infected patients, the impact of co-morbid infections such as HBV and HCV has come into focus. Several studies have monitored HBV or HCV viral loads following initiation of HAART, with disparate results. The effects of HAART on hepatitis B and C plasma viral loads (n = 9 and 32, respectively) and on liver enzyme levels were studied in a large cohort of prospectively studied subjects with advanced stage HIV disease. Comparing the mean pre- and post-HAART levels, there was an estimated increase of (a) 1.40 log(10) from 4.83 to 6.24 log(10) for HBV plasma viral load (P = 0.07), (b) 0.74 log(10) from 6.38 to 7.12 log(10) for HCV plasma viral load (P = 0.001), and (c) 19.4 U/L from 37.4 to 56.8 U/L for serum alanine aminotransferase (P < 0.001). While the number of subjects co-infected with HIV and HBV was limited, these data collected in a population of advanced stage HIV-infected patients raises questions regarding the interactions of these viruses with each other and the host immune system and has implications regarding the order in which antiviral therapies are initiated.

Absence of activation of CMV by blood transfusion to HIV-infected, CMV-seropositive patients.

BACKGROUND: The Viral Activation Transfusion Study (VATS) afforded an opportunity to determine whether blood transfusions, and in particular exogenous WBCs, "activate" CMV replication in HIV-infected, CMV-seropositive patients, and whether such patients can be superinfected by additional strains of CMV transmitted via blood transfusions. STUDY DESIGN AND METHODS: A total of 531 patients were randomized to receive either WBC-reduced (WBCR) or non-WBC-reduced (NWBCR) RBC units. Plasma CMV PCR assays were performed before transfusion and weekly after transfusion for 4 weeks. NWBCR cases with evidence of possible reactivation and/or superinfection were further studied for donor viremia by DNA PCR of frozen retention segments and new genotype acquisition using gB envelope sequence analysis of pre- and posttransfusion recipient specimens. RESULTS: VATS patients received a median of two RBC units during their initial transfusion. Whether positive or negative for CMV DNA at baseline, there were no significant treatment-arm differences in the percentage of patients who had positive qualitative CMV PCR or increases in CMV viral load at follow-up. Of 50 recipients randomized to NWBCR RBC and meeting criteria for possible CMV superinfection, 25 had sufficient CMV DNA load in a baseline and one or more viremic follow-up sample to permit comparison of gB genotypes. Only two recipients showed genotype shifts. Of 125 WBC pellets prepared from the seropositive units transfused into these 50 cases, only 1 tested weakly PCR positive for CMV DNA (insufficient copy number for genotyping). CONCLUSION: There was no evidence of "activation" of CMV by blood transfusion. Among the NWBCR RBC recipients, there was little evidence of possible transmission of new CMV strains. Hence, the current policy for transfusion support of HIV-infected patients, which allows transfusion of CMV-antibody-positive blood to CMV-seropositive patients, is appropriate.

Comparative sensitivity of HBV NATs and HBsAg assays for detection of acute HBV infection.

BACKGROUND: A study was designed to estimate relative analytic sensitivity and window-period (WP) closure and to project incremental yield of newer HBsAg tests, pooled-sample NAT, and single-sample NAT, compared to currently licensed HBsAg tests. STUDY DESIGN AND METHODS: HBV DNA and HBsAg test results for 23 HBV seroconversion (SC) panels were first analyzed to construct a model of primary HBV viremia. One-hundred representative samples were then selected from 10 panels and coded with 28 analytical controls. All 128 samples were tested by seven HBsAg tests and by four pooled-sample and three single-sample NAT assay formats. Results were analyzed to obtain differential times to HBV detection and combined with HBV incidence rates to project comparative yields. RESULTS: HBV doubling time during the ramp-up phase was estimated at 2.56 days. HBsAg concentrations at cutoff for new tests ranged from 0.07 to 0.12 ng per mL, compared with 0.13 to 0.62 ng per mL for licensed tests. Estimated viral load at cutoff ranged from 102 to 267 IU per mL for new tests and from 363 to 1069 IU per mL for licensed tests. HBsAg tests detected 31 to 63 percent of early ramp-up phase samples in the 100-member seroconversion panel study, while pooled-sample NAT detected 55 to 71 percent and single-sample NAT, 82 to 99 percent. Compared with currently licensed HBsAg assays, newer HBsAg assays would reduce the WP by 2 to 9 days; pooled-sample NAT would reduce the WP by 9 to 11 days; and single-sample NAT would reduce the WP by 25 to 36 days. CONCLUSION: Newer HBsAg tests would be expected to detect an additional 15 to 21 infected units per 107 donations, compared to licensed HBsAg tests. Sensitivity, WP closure, and yield projections for newer HBsAg assays and pooled-sample NAT are comparable. Single-sample NAT would increase yield by 13 to 15 units per 107 donations over pooled-sample NAT and newer HBsAg assays and by 35 to 50 units per 107 donations over currently licensed HBsAg assays.

Frequency and duration of plasma CMV viremia in seroconverting blood donors and recipients.

BACKGROUND: Both CMV-seronegative blood and unscreened, filtered blood carry a low but definite risk of transmitting CMV infection. To explain this residual risk, evidence of cell-free viremia was sought in seroconverting and seroprevalent blood donors and seroconverting transfusion recipients by means of a plasma-based assay for CMV DNA. STUDY DESIGN AND METHODS: A CMV DNA PCR assay (COBAS Amplicor CMV Monitor, Roche) was used to detect CMV DNA in 384 paired plasma samples from 192 donors who seroconverted to anti-CMV, 488 anti-CMV EIA-positive samples from 60 seroprevalent donors, and 113 serial samples from 11 seroconverting recipients with posttransfusion CMV hepatitis. RESULTS: Three of 384 samples from 192 seroconverting donors had low levels of plasma CMV DNA (400-1600 copies/mL); one donor was positive before seroconversion, and the other two, after seroconversion. None of the 488 serial samples from 60 anti-CMV- positive donors contained CMV DNA in plasma. Three of 11 recipients demonstrated transient plasma viremia that temporally coincided with seroconversion. CONCLUSIONS: Plasma CMV DNA was detected in a small percentage of seroconverting blood donors and a larger percentage of recipients but was undetectable in seroprevalent donors. Plasma viremia in seroconverting donors may partially explain the low residual risk of CMV transmission by both CMV-seronegative and WBC-reduced seropositive blood.

CMV DNA is rarely detected in healthy blood donors using validated PCR assays.

BACKGROUND: Although serologic screening or WBC reduction of blood components can reduce the incidence of transfusion-transmitted CMV (TT-CMV) infection, 'breakthrough' cases of TT-CMV still occur and may produce serious sequelae. NAT of blood components for CMV DNA has been proposed to further reduce the risks of TT-CMV. However, large-scale studies to determine the utility of validated CMV NAT assays for donor screening have not been reported. STUDY DESIGN AND METHODS: Coded whole-blood samples (n=1000) were tested for the presence of CMV DNA using two CMV PCR assays previously validated in a multicenter trial (a nested PCR assay directed at the CMV UL93 open-reading frame and the Roche Monitor assay). Corresponding plasma samples were tested in parallel for the presence of anti-CMV using other assays (Abbott CMV EIA and Fujirebio/Olympus CMV particle agglutination assays). RESULTS: In total 416 and 514 of the samples tested as CMV-seropositive and -seronegative, respectively, by both antibody assays. The remaining 70 samples had discrepant serology results. Only 2 of the 1000 samples (both seropositive) had reproducibly detectable CMV DNA (positive in at least three of four replicates). CMV DNA was not reproducibly detected in seronegative samples or in samples with discrepant serology results. CONCLUSIONS: Although previous investigations showed frequent detection of CMV DNA in healthy CMV-seropositive (and some seronegative) blood donors, these studies were relatively small and the performance characteristics of their assays were difficult to evaluate. In contrast, the present large cross-sectional study of US donors utilized two previously validated PCR assays and demonstrated that CMV DNA is only rarely detectable in seropositive donors. Thus, the use of CMV PCR assays with optimal performance characteristics did not increase the detection of potentially infectious blood components beyond that provided by current serologic screening assays.

Absence of HBV and HCV, HTLV-I and -II, and human herpes virus-8 activation after allogeneic RBC transfusion in patients with advanced HIV-1 infection.

BACKGROUND: The Viral Activation Transfusion Study was a prospective, randomized, double-blind comparison of transfusion with WBC-reduced versus non-WBC-reduced RBCs to HIV+ patients. The primary study characterized the effect of transfusion on HIV and CMV activation by monitoring viral load changes. The present study analyzed HBV, HCV, HTLV-I and -II, and human herpes virus-8 (HHV-8) viral load before and after transfusion to evaluate the further hypothesis that global immune stimulation following allogeneic RBC transfusion results in activation and increased viral proliferation of chronic viral infections other than HIV and CMV. STUDY DESIGN AND METHODS: Baseline samples from 519 to 523 subjects were screened for HBV, HCV, HTLV-I and -II, and HHV-8 infection, and baseline, serial weekly, and quarterly blood samples from infected subjects in the non-WBC-reduced arm were evaluated for changes from baseline in viral nucleic acid and ALT levels. RESULTS: Seroprevalence of HBV, HCV, HTLV-I and -II, and HHV-8 was 68, 25, 5, and 30 percent, respectively. No significant induction of HBV, HCV, HHV-8, or HTLV-I and -II viral replication following allogeneic transfusion of non-WBC-reduced blood was observed. A significant, albeit small, association was observed between transfusion and ALT. CONCLUSIONS: Based on these results and our previous finding that no adverse effect on HIV and CMV viral load and disease progression results from allogeneic transfusion, no evidence is found to support the selective use of WBC-reduced blood components for HIV-infected patients.

Residual WBC subsets in filtered prestorage RBCs.

BACKGROUND: New-generation RBC filters reduce WBC concentrations by 4 to 5 logs and may prevent or decrease transfusion complications such as HLA alloimmunization, nonhemolytic febrile reaction, and transfusion-transmitted infections. The residual level of WBC subsets may influence efficacy of WBC reduction for preventing various complications. This study analyzed subsets of residual WBCs in WBC-reduced RBC components prepared for a large, multicenter prospective study. STUDY DESIGN AND METHODS: The Viral Activation Transfusion Study (VATS) assessed the impact of WBC reduction in HIV-1-infected patients undergoing RBC transfusion. WBC-reduced RBC from 11 clinical sites with variable filtration practices were sorted into "low,""middle," and "high" groups based on residual WBC concentration. Subsets were isolated from units by immunocapture (anti-CD4-, anti-CD8-, anti-CD15-, and anti-CD19-coated magnetic beads) and quantified by PCR amplification. RESULTS: After validation studies confirming test methodology, 105 VATS WBC-reduced RBC samples were analyzed. Concentrations of subsets in low and middle residual WBC groups were very low in contrast to relatively high concentrations in the high group. Although highly significant differences were identified between the middle and high groups for total WBCs and all subsets, no single subset predominated. CONCLUSION: These results suggest that overall efficacy of WBC filtration correlates with removal of WBC subsets.