Incidence and clinical characteristics of transfusion-associated circulatory overload using an active surveillance algorithm.

BACKGROUND: The concordance of haemovigilance criteria developed for surveillance of transfusion-associated circulatory overload (TACO) with its clinical diagnosis has not been assessed. In a pilot study to evaluate an electronic screening algorithm, we sought to examine TACO incidence and application of haemovigilance criteria in patients with post-transfusion pulmonary oedema. STUDY DESIGN AND METHODS: From June to September 2014, all transfused adult inpatients at four academic hospitals were screened with an algorithm identifying chest radiographs ordered within 12 h of blood component release. Patients with post-transfusion pulmonary oedema underwent case adjudication by an expert panel. TACO incidence was calculated, and clinical characteristics were compared with other causes of post-transfusion pulmonary oedema. RESULTS: Among 4932 transfused patients, there were 3412 algorithm alerts, 50 cases of TACO and 47 other causes of pulmonary oedema. TACO incidence was 1 case per 100 patients transfused. TACO classification based on two sets of haemovigilance criteria (National Healthcare Safety Network and proposed revised International Society for Blood Transfusion) was concordant with expert panel diagnosis in 57% and 54% of reviewed cases, respectively. Although the majority of clinical parameters did not differentiate expert panel adjudicated TACO from other cases, improved oxygenation within 24 h of transfusion did (P = 0·01). CONCLUSIONS: The incidence of TACO was similar to that observed in prior studies utilizing active surveillance. Case classification by haemovigilance criteria was frequently discordant with clinical diagnoses of TACO in patients with post-transfusion pulmonary oedema. Improvements in oxygenation within 24 h of transfusion merit further evaluation in the diagnosis of TACO.

Human immunodeficiency virus infection and indeterminate western blot patterns. Prospective studies in a low prevalence population.

Interpretation of human immunodeficiency virus (HIV) antibody results that are "indeterminate" rather than clearly positive or negative is problematic for the person delivering the result as well as for the individual being tested. To improve counseling messages for these individuals, we evaluated data collected from a well-characterized cohort of 387 blood donors who had been monitored for up to 2 years. We sought to determine if persons with indeterminate Western blot patterns were infected with HIV, and whether information derived from follow-up monitoring would assist in the development of counseling messages for persons on whom no follow-up information was available. Donors were studied by laboratory assays, clinical evaluation, and assessment of risk for HIV. The absence of HIV infection in 97 of 98 donors with indeterminate Western blot patterns was confirmed by clinical follow-up, Western blot assays of sequential samples, and negative gene amplification results. We propose supplemental guidelines to be used as an adjunct to existing interpretive criteria for counseling individuals when they first present with an indeterminate Western blot finding.

Hepatitis B virus (HBV) DNA screening of blood donations in minipools with the COBAS AmpliScreen HBV test.

BACKGROUND: The risk of hepatitis B virus (HBV) transmission by blood transfusion (estimated at 1 in 63,000-1 in 205,000 units in the United States) exceeds that of hepatitis C virus (HCV) or human immunodeficiency virus (HIV). Reduction of window-period HBV transmissions through detection of HBV DNA-positive units by minipool nucleic acid testing (MP NAT) would be expected to decrease this risk. STUDY DESIGN AND METHODS: A large multicenter study of the COBAS AmpliScreen HBV test (Roche Molecular Systems) was conducted on minipools of 24 blood donation specimens. The yield of HBV DNA-positive, hepatitis B surface antigen (HBsAg)-negative window-period donations was determined relative to current and newly licensed HBsAg assays. Donors with selected HBV DNA, HBsAg, and anti-hepatitis B core antigen (HBc) results were further evaluated. RESULTS: The detection rate of window-period units was 1 in 352,451 (95% confidence interval, 1 in 2,941,176-1 in 97,561). Assay specificity was high (99.9964%). HBV DNA was detected in 84 percent of HBsAg-positive, anti-HBc-positive donations by MP NAT and in 94 percent when individual-donation (ID) NAT was added. HBV DNA was detected in 0.03 percent of HBsAg-negative, anti-HBc-positive donations by MP NAT and in 0.41 percent when ID NAT was added. CONCLUSIONS: Implementation of HBV MP NAT will provide an increment in safety relative to HBV serologic screening, similar to that for HCV and in excess of that for HIV. Our data indicate that the implementation of HBV MP NAT would likely interdict 39 HBV window-period units and prevent 56 cases of transfusion-transmitted HBV infection annually. The current data indicate that HBV MP NAT should not lead to discontinuation of anti-HBc testing but that discontinuation of HBsAg testing with retention of anti-HBc testing may be possible.

The RADAR repository: a resource for studies of infectious agents and their transmissibility by transfusion.

BACKGROUND: An ongoing issue in transfusion medicine is whether newly identified or emerging pathogens can be transmitted by transfusion. One method to study this question is through the use of a contemporary linked donor-recipient repository. STUDY DESIGN AND METHODS: The Retrovirus Epidemiology Donor Study Allogeneic Donor and Recipient (RADAR) repository was established between 2000 and 2003 by seven blood centers and eight collaborating hospitals. Specimens from consented donors were collected, components from their donations were routed to participating hospitals, and recipients of these units gave enrollment and follow-up specimens for long-term storage. The repository was designed to show that zero transmissions to enrolled recipients would indicate with 95 percent confidence that the transfusion transmission rate of an agent with prevalence of 0.05 to 1 percent was lower than 25 percent. RESULTS: The repository contains pre- and posttransfusion specimens from 3,575 cardiac, vascular, and orthopedic surgery patients, linked to 13,201 donation specimens. The mean number of RADAR donation exposures per recipient is 3.85. The distribution of components transfused is 77 percent red cells, 13 percent whole blood-derived platelet concentrates, and 10 percent fresh frozen plasma. A supplementary unlinked donation repository containing 99,906 specimens from 84,339 donors was also established and can be used to evaluate the prevalence of an agent and validate assay(s) performance before accessing the donor-recipient-linked repository. Recipient testing conducted during the establishment of RADAR revealed no transmissions of human immunodeficiency virus, hepatitis C virus, or human T-lymphotropic virus. CONCLUSIONS: RADAR is a contemporary donor-recipient repository that can be accessed to study the transfusion transmissibility of emerging agents.

Detection of West Nile virus RNA and antibody in frozen plasma components from a voluntary market withdrawal during the 2002 peak epidemic.

BACKGROUND: The US West Nile virus (WNV) epidemic in the summer and fall of 2002 included the first documented cases of transfusion-transmitted WNV infection. In December 2002, the FDA supported a voluntary market withdrawal by the blood banking community of frozen blood components collected in WNV high-activity areas. At the time, the prevalence of viremia and serologic markers for WNV in the blood supply was undefined. STUDY DESIGN AND METHODS: In collaboration with America's Blood Centers, 1468 frozen plasma components (of approx. 60,000 frozen units voluntarily withdrawn from the market) were selectively retrieved from the peak epidemic regions and season (June 23, 2002-September 28, 2002). These units were unlinked, subaliquoted, and tested by WNV enzyme immunoassays (EIAs; Focus Technologies and Abbott Laboratories) and nucleic acid amplification tests (NATs; Gen-Probe Inc. and Roche Molecular Systems). RESULTS: Of the 1468 EIA results from Abbott and Focus, 7 were anti-immunoglobulin M (IgM)- and anti-immunoglobulin G (IgG)-reactive by both assays, 8 and 1 were IgM-only-reactive, and 8 and 23 were IgG-only-reactive, respectively. NAT by Gen-Probe and Roche Molecular Systems yielded one RNA-positive, antibody-negative unit containing approximately 440 RNA copies per mL. An additional 10-fold replicate NAT testing by Gen-Probe on 14 of 15 IgM-reactive specimens yielded 2 additional IgM- and IgG-reactive units with low-level viremia (i.e., 7/10 and 2/10 replicates tested reactive). CONCLUSION: The prevalence of acute (RNA-positive) and recent (IgM-seroreactive) WNV infections indicates that transfusion risk in high-risk areas could have been considerable and that voluntary market withdrawal of frozen components likely averted some WNV transfusion transmissions. The existence of very-low-level viremic units raises concerns, because WNV minipool NAT screening will miss such units and individual NAT may not completely correct this situation.

Analytical and clinical sensitivity of West Nile virus RNA screening and supplemental assays available in 2003.

BACKGROUND: Transfusion-transmitted West Nile virus (WNV) infections were first reported in 2002, which led to rapid development of investigational nucleic acid amplification tests (NAT). A study was conducted to evaluate sensitivities of WNV screening and supplemental NAT assays first employed in 2003. STUDY DESIGN AND METHODS: Twenty-five member-coded panels were distributed to NAT assay manufacturers. Panels included five pedigreed WNV standards (1, 3, 10, 30, and 100 copies/mL), 15 or 16 donor units with very-low-level viremia identified through 2003 screening, and four or five negative control samples. Samples were tested neat in 10 replicates by all assays; for NAT screening assays, 10 replicates were also performed on dilutions consistent with minipool (MP)-NAT. The viral load distribution for 142 MP-NAT yield donations was characterized, relative to the analytical sensitivity of MP-NAT systems. RESULTS: Analytical sensitivities (50% limits of detection [LoD] based on Poisson model of detection of WNV standards) for screening NAT assays ranged from 3.4 to 29 copies per mL; when diluted consistent with MP pool sizes, the 50 percent LoD of screening NAT assays was reduced to 43 to 309 copies per mL. Analytical sensitivity of supplemental assays ranged from 1.5 to 7.7 copies per mL (50% LoD). Detection of RNA in donor units varied consistent with analytical LoD of assays. Detection of low-level viremia after MP dilutions was particularly compromised for seropositive units, probably reflecting lower viral loads in the postseroconversion phase. Based on the viral load distribution of MP-NAT yield donations (median, 3519 copies/mL; range, < 50-690,000), 13 to 24 percent of units had viral loads below the 50 percent LoD of screening NAT assays run in MP-NAT format. CONCLUSION: WNV screening and supplemental assays had generally excellent analytical sensitivity, comparable to human immunodeficiency virus-1 and hepatitis C virus NAT assays. The presence of low-level viremic units during epidemic periods and the impact of MP dilutions on sensitivity, however, suggest the need for further improvements in sensitivity as well as a role for targeted individual-donation NAT in epidemic regions.

The risks of transfusion-transmitted infection: direct estimation and mathematical modelling.

Direct measurement of the risk of transfusion-transmitted infection (TTI) is practical and accurate only if the level of risk is high. Historically, studies that established frozen repositories of transfusion recipient and/or blood donor samples were important in establishing the risk of many TTI agents, including the human immunodeficiency virus (HIV), hepatitis B virus (HBV) and hepatitis C virus (HCV). However, given the current very low risk of TTI, mathematical modelling is necessary to estimate the magnitude of such a risk. For agents for which routine blood donor screening is performed, most of this risk comes from transfusion of units collected in the window period between donor infection and a positive blood screening assay. The incidence/window period model has been used to estimate the magnitude of such risks (of the order of 1:100 000 to 1:1 000 000) and for predicting the extent of risk reduction that can be expected with implementation of new tests. Direct estimation and mathematical modelling approaches are both important tools for future assessment of potential, new or emerging TTI agents.

Use of erythrocytapheresis in the treatment of patients with sickle cell anemia.

We performed exchange transfusions, utilizing the technique of automated erythrocytapheresis, for the treatment of patients with sickle cell anemia. In an attempt to determine guidelines for the use of erythrocytapheresis, we studied the use of this procedure in three distinct clinical situations in nine patients with sickle cell disease. Patients with dangerous complications of sickle cell disease such as acute respiratory distress and priapism responded well to erythrocytapheresis, showing marked improvement within 24-48 hours. Patients with prolonged painful vasoocclusive crises showed only variable improvement after erythrocytapheresis therapy, insufficient to justify exposing the patient to the risks of the procedure. Patients treated to decrease the frequency of painful crises demonstrated no prolongation in symptom-free intervals between crises. Therefore, erythrocytapheresis has its main value in the management of acute, dangerous complications of sickle cell disease.

CD4 and CD8 subsets defined by dual-color cytofluorometry which distinguish symptomatic from asymptomatic blood donors seropositive for human immunodeficiency virus.

As part of a larger study to characterize immune alterations in blood donors seropositive for human immunodeficiency virus (HIV), we measured subsets of CD4 (T helper/inducer) and CD8 (T suppressor/cytotoxic) cells by 2-color cytofluorometry. Alterations observed in asymptomatic seropositive donors (ASP) included: 1) decreased mean levels of Leu 8+ CD4 cells, although the proportion of Leu 8+ cells within the CD4 population was unchanged; 2) a selective increase in Leu 8- CD8 and Leu 18- CD8 cell levels; and 3) increased levels of both CD8 subsets defined by Leu 7, Leu 17, or HLA-DR expression. Alterations observed in symptomatic seropositive donors (SSP) were: 1) a further decrease in Leu 8+ CD4 cell levels, with a decrease in the proportion of Leu 8+ CD4 cells; 2) decreased levels of both Leu 18- and Leu 18+ CD4 subsets; 3) selective increases in Leu 8- and Leu 18- CD8 cell levels; and 4) increases in Leu 7+ CD8, Leu 17+ CD8, and HLA-DR+ CD8 subsets but not the reciprocal negative CD8 subsets. Thus, changes merely reflective of HIV infection included decreased levels of Leu 8+ CD4 cells and increased levels of Leu 8- CD8, Leu 18- CD8, Leu 7+ CD8, Leu 17+ CD8, and HLA-DR+ CD8 cells. Development of symptoms were associated with a further, preferential loss of Leu 8+ CD4 cells, proportional losses of both CD4 subsets defined by Leu 18 expression, and a return to normal levels of Leu 7- CD8, Leu 17- CD8, and HLA-DR- CD8 cells.

In vitro activation of T lymphocytes from human immunodeficiency virus (HIV)-seropositive blood donors. I. Soluble interleukin 2 receptor (IL2R) production parallels cellular IL2R expression and DNA synthesis.

We investigated the relationship of soluble interleukin 2 receptor (sIL2R) production to cellular IL2R expression and DNA synthesis by mitogen-stimulated mononuclear cells from blood donors seropositive for human immunodeficiency virus (HIV). SIL2R was measured using an enzyme-linked immunosorbent assay which employed 2 anti-IL2R monoclonal antibodies recognizing distinct IL2R epitopes. Decreased phytohemagglutinin-induced DNA synthesis and cellular IL2R expression were accompanied by decreased levels of sIL2R in cell culture supernatants. Similar findings were observed for pokeweed mitogen-induced responses. There was no detectable spontaneous secretion of sIL2R into culture supernatants by unstimulated mononuclear cells from either HIV-seropositive or control seronegative donors. These findings indicate that the in vitro T-cell activation defects which characterize HIV infection include decreased sIL2R production, as well as decreased cellular IL2R expression and DNA synthesis. Further, they show that assessment of supernatant sIL2R levels can be used as a valid, reliable assay for T-cell activation.

The risk of transfusion-transmitted viral infections. The Retrovirus Epidemiology Donor Study.

BACKGROUND: Accurate estimates of the risk of transfusion-transmitted infectious disease are essential for monitoring the safety of the blood supply and evaluating the potential effect of new screening tests. We estimated the risk of transmitting the human Immuno-deficiency virus (HIV), the human T-cell lymphotropic virus (HTLV), the hepatitis C virus (HCV), and the hepatitis B virus (HBV) from screened blood units donated during the window period following a recent, undetected infection. METHODS: Using data on 586,507 persons who each donated blood more than once between 1991 and 1993 at five blood centers (for a total of 2,318,356 allogeneic blood donations), we calculated the incidence rates of seroconversion among those whose donations passed all the screening tests used. We adjusted these rates for the estimated duration of the infectious window period for each virus. We then estimated the further reductions in risk that would result from the use of new and more sensitive viral-antigen or nucleic acid screening tests. RESULTS: Among donors whose units passed all screening tests, the risks of giving blood during an infectious window period were estimated as follows: for HIV, 1 in 493,000 (95 percent confidence interval, 202,000 to 2,778,000); for HTLV, 1 in 641,000 (256,000 to 2,000,000); for HCV, 1 in 103,000 (28,000 to 288,000); and for HBV, 1 in 63,000 (31,000 to 147,000). HBV and HCV accounted for 88 percent of the aggregate risk of 1 in 34,000. New screening tests that shorten the window periods for the four viruses should reduce the risks by 27 to 72 percent. CONCLUSIONS: The risk of transmitting HIV, HTLV, HCV, or HBV infection by the transfusion of screened blood is very small, and new screening tests will reduce the risk even further.

Evaluation of a p21e-spiked western blot (immunoblot) in confirming human T-cell lymphotropic virus type I or II infection in volunteer blood donors. The Retrovirus Epidemiology Donor Study Group.

Current algorithms for the serologic confirmation of human T-cell lymphotropic virus type I and II (HTLV-I/II) antibody reactivity are complicated. We evaluated the performance of an HTLV-I Western blot (immunoblot) spiked with recombinant p21e protein (p21e WB) as an alternative to current confirmatory methods. These methods include the HTLV-I viral lysate Western blot and either a radioimmunoprecipitation assay or a p21e enzyme-linked immunosorbent assay. Five hundred fifty nine blood donations obtained from five U.S. blood centers and classified as HTLV-I/II seropositive (n = 149) or seroindeterminate (n = 410) by routine testing methods were further evaluated by PCR for proviral DNA and by the p21e WB. On the basis of serologic and PCR testing, 155 donations were classified as HTLV-I/II infected. The sensitivity of the p21e WB was 97.4%, slightly exceeding that of routine confirmatory testing. The specificity of the p21e WB was 97.5%, as determined by testing of 404 seroindeterminate samples that were negative in the PCR. The positive predictive value of the p21e WB was 94%. In contrast, the specificity and positive predictive value of routine confirmatory testing were both 100%. Follow-up sampling of presumptive p21e WB false-positive donors substantiated the absence of HTLV-I/II infection. Although the p21e WB used in this study has high sensitivity and may be useful as a confirmatory assay in epidemiologic research studies, it may not be ideal as a confirmatory test for the notification of blood donors.

Epidemiologic characteristics of blood donors with antibody to human immunodeficiency virus.

From March 1985 through July 1986, blood donors who were positive for antibody to human immunodeficiency virus (HIV) were evaluated at three major blood centers in the United States. Of 818,629 donations, 450 (0.05%) were HIV antibody-positive. The seroprevalence decreased from 0.07 to 0.04 percent during the study period, due perhaps to a decline in repeat donors. HIV-seropositive donors tended to be 20 to 29 years old (52%) and male (88%). HIV seroprevalence among white donors (2/10,000 donations) was less than that among Hispanic (9/10,000; p less than 0.0001) and black donors (31/10,000; p less than 0.0001). Of 152 seropositive men interviewed, 77 percent reported sexual contact with men; of this latter group, 53 percent were bisexual. Fifteen (44%) of 34 seropositive women had apparently acquired infection from heterosexual contact, and an equal number denied having any known risk factors for HIV infection. Educational efforts must address women and bisexual men who do not perceive themselves to be at risk for HIV infection and should be specifically designed for the mores of different racial and ethnic groups.

Declining value of alanine aminotransferase in screening of blood donors to prevent posttransfusion hepatitis B and C virus infection. The Retrovirus Epidemiology Donor Study.

BACKGROUND: Since the mid-1980s, blood banks in the United States have screened donors for elevated alanine aminotransferase (ALT) in an effort to prevent posttransfusion hepatitis. The present study was designed to quantitate the residual value of ALT screening following the implementation of hepatitis C virus (HCV) assays. STUDY DESIGN AND METHODS: Two approaches were used. First, a database of 2.3 million donations made by 586,507 volunteer blood donors between 1991 and 1993 was used to compare the incidence of seroconversion to hepatitis B virus (HBV) and HCV marker positivity in donors with elevated ALT values and with normal ALT values. Second, the duration of ALT elevation prior to HBV and HCV seroconversion was determined from 34 well-documented cases of posttransfusion HBV and HCV; elevated-ALT window periods were multiplied by rates of HBV and HCV incidence in donors to project the yield of ALT screening. Predictive value and cost-effectiveness analyses were also performed to compare the value of ALT screening before and after HCV screening was implemented. RESULTS: Both approaches indicate that ALT testing does not detect HBV in the window phase but does currently identify approximately 3 HCV window-phase donations per 1 million donations; this contrasts with ALT detection of approximately 1800 HCV-infectious units per 1 million donations prior to anti-HCV screening. Currently, only 8 in 10,000 donated units with elevated ALT (negative anti-HCV) are infected with HCV. The cost of continued ALT screening was estimated at $7,931,000 per quality-adjusted year of life saved. CONCLUSION: The yield, predictive value, and cost-effectiveness of ALT screening of blood donors have declined dramatically with the implementation of progressively improved anti-HCV assays. ALT screening of volunteer blood donors should be discontinued.

Incidence rates of viral infections among repeat donors:are frequent donors safer?

BACKGROUND: Incidence rates (IRs) for viral infections may vary with the frequency of donation among repeat, community, whole-blood (WB) donors, with IRs thought to be lower among donors with higher frequency of donation. STUDY DESIGN AND METHODS: IRs for HIV, HTLV, HCV, and HBV infection were stratified by frequency of donation among 868,403 repeat WB donors who gave approximately 4 million donations at five United States blood centers from 1991 through 96. All donors had given at least 2 donations during those years, with the first donation being nonreactive on confirmatory testing. Frequency of donation was measured in three ways: by the number of donations per year; at the time of donation, by the number of donations given within the preceding 2-year period; and by the number of donations given from 1991 through 1993. RESULTS: The IRs for HIV, HCV, and HBV infection did not appear to differ among donors with lower or higher numbers of donation per year. However, the IR for HTLV infection decreased as the number of donations per year increased (p = 0.0004). IRs for all viral markers remained stable, regardless of the number of donations given within the 2-year period before the donation. Although IRs for HIV, HTLV, and HCV infection did not vary by the number of donations given in 1991 through 1993, the IR for HBV infection appeared to be lower in donors who gave more donations in that period (p = 0.01). CONCLUSION: These findings do not provide evidence of a lower IR for transfusion-transmissible viral infections among repeat WB donors who give more frequently. Abbreviated screening histories for frequent repeat donors might not be advisable.

Trends in incidence and prevalence of major transfusion-transmissible viral infections in US blood donors, 1991 to 1996. Retrovirus Epidemiology Donor Study (REDS)

CONTEXT: Evaluating trends in blood donor infectious disease rates is essential for monitoring blood supply safety and donor screening effectiveness. OBJECTIVE: To determine changes over time in blood donor population infection rates of human immunodeficiency virus (HIV), human T-lymphotropic virus (HTLV), hepatitis C virus (HCV), and hepatitis B virus (HBV). DESIGN: Cross-sectional survey data from the National Heart, Lung, and Blood Institute-sponsored Retrovirus Epidemiology Donor Study. SETTING: Five blood centers in different regions of the United States. PARTICIPANTS: A total of 1.9 million volunteer blood donors with 1 or more nonautologous donations from January 1991 to December 1996. MAIN OUTCOME MEASURES: Changes in rates of HIV, HTLV, HCV, and HBV infections were evaluated by comparing yearly prevalence estimates (per 100,000 donations) for first-time allogeneic donors and period-specific incidence rates (IRs) (per 100,000 person-years) for repeat allogeneic donors between 1991 and 1996 (for HCV, from about March 1992 to June 1996). RESULTS: Prevalence of HIV decreased in first-time donors from 0.030% to 0.015% (P=.006) and HCV prevalence decreased from 0.63% to 0.40% (P<.001). Trends were not statistically significant for the proportion of first-time donors with hepatitis B surface antigen (HBsAg) or HTLV. For repeat donors, IRs did not change significantly, indicating a stable but low level of seroconversion. The overall IRs (95% confidence intervals) per 100,000 person-years were 2.92 (2.26-3.70) for HIV, 1.59 (1.12-2.19) for HTLV, 3.25 (2.36-4.36) for HCV, and an estimated 10.43 (7.99-13. 37) for HBV (based on an HBsAg rate of 2.66 [2.04-3.41] with presumed false-positive results considered negative). The HBV IR estimate with presumed false-positive results considered positive (for comparability to previous analyses) was 17.83 (14.60-21.56). CONCLUSION: The decrease in HIV and HCV prevalence rates, combined with the previously documented lower rates of infection in first-time donors compared with the general population, suggests the continued benefit of behavioral risk factor screening. JAMA. 2000;284:229-235

Loss of volunteer blood donors because of unconfirmed enzyme immunoassay screening results. Retrovirus Epidemiology Donor Study.

BACKGROUND: Blood donors who test repeatably reactive on enzyme immunoassay (EIA) and are not confirmed as positive are a continuing problem for blood banks. Units are discarded and donors are deferred, in spite of multiple studies indicating that such donors are very rarely infected with the transmissible agents. Few data are available, however, with which to evaluate whether the discarded units are more likely to come from particular demographic subgroups. STUDY DESIGN AND METHODS: The Retrovirus Epidemiology Donor Study database of over 2 million allogeneic whole-blood donations collected in the years 1991 through 1993 was utilized. The prevalence of false-positive and indeterminate test results within demographic subgroups was computed for antibodies to human immunodeficiency virus, hepatitis C virus, and human T-lymphotropic virus (anti-HIV, anti-HCV, anti-HTLV, respectively) and hepatitis B surface antigen (false-positive only) as the proportion of donations that were repeatably reactive on EIA but negative or indeterminate on the confirmatory or supplemental test. RESULTS: Several demographic groups with increased prevalence of false-positive and indeterminate anti-HIV results were the same females, younger age groups, blacks, and first-time donors. Likewise, many of the demographic subgroups with increased prevalence of false-positive and indeterminate anti-HCV results were similar: older age groups, non-whites, lower education levels, first-time donors, donors making directed donations, and donors who had received transfusions. For anti-HTLV, by contrast, the oldest group had the highest prevalence of false-positive results but the lowest prevalence of indeterminate results: blacks had the lowest prevalence of false positive results but the highest prevalence of indeterminate results. CONCLUSION: If units that test repeatably reactive on EIA but that are not confirmed as positive are almost always from individuals not infected with the virus in question, then these results indicate that there may be sex-, race-, and/or age-linked proteins cross-reacting with the test kit materials. Elucidation of these antigenic determinates and their subsequent removal should be a priority.

Accuracy of supplementary serologic testing for human T-lymphotropic virus types I and II in US blood donors. Retrovirus Epidemiology Donor Study.

Blood donations in the United States have been screened for antibody to human T-lymphotropic virus type I (HTLV-I) by HTLV-I enzyme immunoassay (EIA) since November 1988. Specimens repeatedly found to be reactive by EIA undergo confirmation by supplementary serologic tests. We assessed the accuracy of blood center testing of 994 HTLV-I EIA repeat-reactive specimens in five US blood centers between November 1988 and December 1991. Of 410 confirmed HTLV-I/II donations, 407 (99.3%) were infected with HTLV-I/II, as determined by polymerase chain reaction (PCR) (403 cases) and by repeat serologic testing (4 cases). The three false-positive results occurred in the first year of testing. Of 425 HTLV-indeterminate specimens, 6 (1.4%) were found to be infected by PCR (5 with HTLV-II and 1 with HTLV-I). None of 159 confirmatory test-negative donations was PCR positive. Of HTLV-I/II-seropositive specimens, 80.2% to 95.4% could be typed as HTLV-I or HTLV-II by type-specific serologic assays. These results support recommendations that HTLV-I/II-seropositive donors should be advised that they are infected with HTLV-I, HTLV-II, or HTLV-I/II (depending on results of type-specific assays). HTLV-indeterminate donors should be advised that their results only rarely indicate HTLV infection. HTLV confirmatory test-negative donors should be reassured that they are not infected with HTLV-I or HTLV-II.