Sensitivity and specificity of commercial ELISA kits for screening anti-LAV/HTLV III

Two panels consisting of 236 and 258 anti-LAV/HTLV III positive and 64 and 50 negative sera, respectively (determined in Western blots), were tested with 8 different, commercially available ELISA test kits for detecting LAV/HTLV III antibodies (Abbott, Dupont, Electro-Nucleonics, Litton, Organon, Pasteur, Sorin, Wellcome). Positive sera were selected for levels of antibodies ranging from average to very high, and the negative sera included both negative sera and sera with antibodies to cellular components such as HLA antigens or nuclear membranes. In examination of the 2 panels, the sensitivity of the tests ranged from 97.2 to 100%, and the specificity from 70.0 to 100%. No test was ideal.     

Evaluation of a new assay for antibodies to LAV/HTLV III

The sensitivity, specificity and reproducibility of an enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies to LAV/HTLV III produced by Genetic Systems was assessed with the identical panel of sera used in previous evaluations of anti-HTLV III ELISAs. The results from this study show that the Genetic Systems anti-LAV/HTLV III ELISA proved to be of equivalent sensitivity and to have higher specificity than assays currently used in Australia for screening purposes while maintaining high levels of intra- and inter-laboratory reproducibility.     

Clinical evaluation of Abbott and Wellcome enzyme linked immunosorbent assays for detection of serum antibodies to human immunodeficiency virus (HIV).

Abbott and Wellcome enzyme linked immunosorbent assays (ELISAs) for detection of antibodies to human immunodeficiency virus (HIV) were compared in tests on 932 sera collected predominantly from male homosexuals attending a sexually transmitted disease (STD) clinic in central London. Two hundred and twenty three sera had HIV antibodies detected by both types of assay, with confirmation of the results by further tests carried out at the Virus Reference Laboratory (VRL) in Colindale. There was a 97.3% correlation between the results obtained by the two commercial ELISA assays on the tests carried out on unheated sera. The Abbott ELISA gave significantly more false positive results than the Wellcome test when the manufacturer's instructions for cut off values were followed. There was one false negative Abbott results: it failed to react to repeated Abbott ELISA but was positive by Wellcome and confirmatory assays. Of 283 heat treated sera 14.8% gave falsely reactive results with the Abbott assay whereas there were no differences between heated and unheated sera with the Wellcome assay. VRL or Western blot confirmatory assays, or both, confirmed all the 235 positive results obtained with the Wellcome assay.     

Assay of digoxin on dry-reagent strips

An automated immunoassay for Digoxin in serum using dry-reagent strips (Stratus Dade) was evaluated, and compared with two others immunoassays (TDX Abbott and ACA Du Pont). Precision, sensitivity and specificity were satisfying, as well as correlation study, but discrepancies observed for some sera proved that these techniques still pose serious problems of specificity or of standardisation.     

Blind comparison of Abbott and Dupont HIV antigen ELISA tests for detecting antigenaemia in asymptomatic human immunodeficiency virus antibody positive homosexual men.

Three hundred and ninety eight serum samples from 270 human immunodeficiency virus (HIV) antibody positive asymptomatic homosexual men were tested in the Abbott and Dupont HIV antigen ELISA tests. In the Abbott test 62 (16%) of the sera were positive, according to the manufacturer's instructions, compared with 55 (14%) in the Dupont test. Twenty six sera were positive with the Abbott test but negative with the Dupont test and 19 sera were positive only by the Dupont test. Only 36 (9%) of the sera were positive in both tests. The Abbott confirmatory neutralisation test gave excellent agreement with the initial Abbott HIV antigen ELISA test; the Dupont confirmatory test was only in agreement with the initial positive Dupont antigen ELISA test in one third of the sera tested. Although the overall sensitivity of each of the two commercial assays tested was similar, the Abbott method may be preferable for clinical purposes if confirmation of an initial ELISA positive test result by neutralisation assay is required.     

Evaluation of the INNO-LIA HTLV I/II assay for confirmation of human T-cell leukemia virus-reactive sera in blood bank donations

We have evaluated a new serological confirmatory test (INNO-LIA HTLV I/II Ab [INNO-LIA]) for human T-cell leukemia virus (HTLV) using a large collection of samples from Brazilian blood donors (S?o Paulo region) and compared the results with those obtained by Western blotting (WB) tests (WB2.3 and WB2.4). Blood donations were initially screened by enzyme-linked immunosorbent assays (ELISAs) based on viral lysates, and repeatedly reactive samples were further tested by WB2.3. When available, samples were also tested by PCR, two additional ELISAs based on recombinant antigens (recombinant ELISAs), a new-generation WB assay (WB2.4), and the INNO-LIA. Of the 18,169 samples tested, 292 (1.61%) were repeatedly reactive in the ELISAs (viral lysate based) and were further tested by WB2.3; 97 were positive (19 that were typed as HTLV type I [HTLV-I], 12 that were typed as HTLV type II [HTLV-II], and 66 that were nontypeable), 17 were negative, and 178 had indeterminate results. Of the samples with indeterminate results, 172 were tested by INNO-LIA, which could resolve 153 samples as negative. Regarding the positive samples, WB2. 3 and INNO-LIA produced concordant results for all HTLV-I-positive samples, whereas for HTLV-II they agreed for 10 of 12 samples; the 2 samples with discordant results were considered to be positive for HTLV-II by WB with WB2.3 but negative for HTLV-II by INNO-LIA and the two recombinant ELISAs. Furthermore, of the 66 nontypeable samples, 60 underwent testing by INNO-LIA; 54 turned out to be negative by the latter test as well as by recombinant ELISAs. In conclusion, the new serological confirmatory assay for HTLV (INNO-LIA HTLV I/II Ab) resolved the results for the majority of the indeterminate and positive-untypeable samples frequently observed by WB assays.     

Identification of HTLV p19 specific natural human antibodies by competition with monoclonal antibody

A competitive binding assay using a monoclonal antibody to the human T-cell lymphoma/leukemia virus (HTLV) p19 was developed for use in detecting natural antibodies to the protein in human sera. The specificity of the assay for HTLV p19 was demonstrated using a variety of antisera. While sera known to contain antibodies to HTLV p19 competed in the assay, antisera prepared against purified HTLV p24, the major core protein of the virus, or against other disrupted type-C retroviruses did not. Sera of Japanese patients with adult T-cell leukemia and similar T-cell malignant lymphomas were examined by this technique for the presence of antibodies to HTLV p19. The results were compared with those obtained by a solid-phase radioimmunoassay (RIA) against disrupted HTLV. The majority of Japanese ATL patients possess natural antibodies to HTLV as shown by solid-phase RIA (88%) and also specifically to HTLV p19 (77%). Similarly, 50% of Japanese patients with similar T-cell malignant lymphomas possess HTLV antibodies by solid-phase RIA and nearly as many (42%) possess anti-p19 reactivity. Twelve and eight percent, respectively, of normal Japanese donors from the ATL endemic region possessed HTLV-specific antibody by the solid-phase RIA or competitive binding assay. Normal donors from nonendemic areas lacked antibodies to HTLV. These results extend our previous findings of natural antibodies to HTLV in Japanese patients with ATL. The finding of p19-specific antibodies in these Japanese sera, together with previous reports of natural antibodies to HTLV p24 in sera from this same geographic cluster, strengthens the association of HTLV with Japanese ATL.     

Evaluation of viral-lysate enzyme-linked immunosorbent assay kits for detecting human immunodeficiency virus (type 1) infections using human sera standardized by quantitative western blotting.

The first generation of proprietary reagents for detecting antibodies to the Human Immunodeficiency Virus Type 1 (HIV-1) by enzyme-linked immunosorbent assay (ELISA) used as antigen partially purified virus from cell culture lysates. These tests, which are still in use, may vary in their antibody measurement capabilities if different proportions of the viral polypeptides are present in the viral lysate mixtures. We determined the quantities of antibodies in the serum of persons infected with HIV-1 by dilution analysis using 3 ELISA kits: Abbott [A], Du Pont [D], Genetic Systems [G]. The proportionate antibody titres of each serum to p24gag and gp160env/120env were established by quantitative Western blotting. Serum antibody titres were high, frequently over 1:10,000, a result observed both by ELISA and Western blot. For Kit D, sera with high proportions of antibody to p24gag produced antibody titration curves with steep slopes whereas shallower slopes were found in sera with high proportions of antibody to gp160env. In contrast, Kit A gave steeper slopes with sera enriched for gp160env antibodies. Kit G gave results with slopes intermediate between Kits A and D. Serum antibody titres differed between kits depending upon the proportion and concentration of antibodies in a given serum to gp160env and p24gag. The findings that both the concentration and proportion of antibodies to specific viral polypeptides in human sera markedly affect the signal intensity produced by proprietary ELISAs suggest the need for several control sera which reflect the diversity of human serum responses. Standardization of human reference sera by quantitative Western blotting will assist in evaluation and quality control of ELISA tests.     

Comparative study of 2 commercial tests for the detection of HIV-1 antigens in blood and cerebrospinal fluid using immunoenzymology

The authors have carried out, on 150 sera of patients seropositive for the human immunodeficiency virus type I (HIV I) and 11 cerebrospinal fluid of which 5 were patient infected by the HIV I, a comparative study of two commercial tests for the detection of HIV I antigen (Diagnostic Pasteur and Abbott laboratories). A much greater sensitivity was obtained with the specificity being practically identical for the sera with the two tests (100% with Abbott laboratories test, 96.11% with the diagnostic Pasteur test). 4 sera appeared "false negatives" with the Abbott Laboratories test; their optical density was situated between 80 and 100 p. cent of the cut-off level value, whereas that of the "real" negatives was situated between 30 and 60 p. cent of the cut-off level value. 10 of the 11 cerebrospinal fluids appeared false positive with the Diagnostic Pasteur. This seems to be connected with an insufficiency of saturation of protein receptors in the wells. The Diagnostic Pasteur test is not adapted for the detection of HIV I antigen in the body fluids with a weak protein concentration. Contrary to the results obtained with the Encavor test (Abbott laboratories) the analysis in western-blot does not show an inverse prevalence of anti p24 GAG antibodies with regard to antigen HIV I in seropositive patients. On the other hand, the statistical analysis of the positive HIV I sera which are at the same time antigen HIV I positive and antibodies HIV I positive suggests an earlier disappearance of anti p17 GAG antibodies than of anti p24 GAG antibodies.     

Reliable confirmation of antibodies to human immunodeficiency virus type 1 (HIV-1) with an enzyme-linked immunoassay using recombinant antigens derived from the HIV-1 gag, pol, and env genes.

An enzyme-linked immunoassay (ELISA) using six recombinant proteins corresponding to large segments of the human immunodeficiency virus type 1 (HIV-1) gag, pol, and env gene products (HIVAGEN; SmithKline Bio-Science Laboratories, Van Nuys, Calif.) was developed to confirm the presence of antibodies to HIV-1 in sera reactive in the whole-cell-derived virion screening ELISAs. Serum samples for testing were obtained from healthy seronegative blood donors and from the different categories of HIV-infected individuals (asymptomatic, acquired immunodeficiency syndrome [AIDS]-related complex, and AIDS). A positive reaction was defined as reactivity against an env and at least one other (either gag or pol) HIV-1 gene product; negative was defined as no reaction with any antigen; and indeterminate was defined as reactivity with gag or pol (or both) or with env alone. None of the 1,180 serum samples from healthy seronegative blood donors gave a positive result, and only 49 of these samples (4%) gave indeterminate results. The recombinant HIV-1 antigen ELISA panel identified seropositive individuals with a high degree of accuracy, as a positive reaction was seen with 99.3% of asymptomatic healthy seropositive individuals, 98.1% of patients with AIDS-related complex, and 90.4% of patients with AIDS. None of the 725 HIV-1-seropositive subjects had a negative test result. Reactivity with the Kp41 antigen, corresponding to an amino-terminal portion of the gp41 envelope glycoprotein, by itself demonstrated 100% sensitivity and specificity in distinguishing seronegative from seropositive sera. A subset of seronegative and seropositive samples were tested both with the recombinant HIV-1 antigen ELISA panel and by Western blot (Du Pont Co.). The recombinant HIV-1 antigen ELISA panel accurately identified more seropositive and seronegative samples and had fewer indeterminate results than did Western blot (interpreted by Du Pont criteria).     

A comparison between one first generation and three second generation anti-HCV ELISAs: an investigation in high- and low-risk subjects in correlation with recombinant immunoblot assay and polymerase chain reaction

One first generation assay (manufactured by Ortho, test I) and 3 second generation anti-HCV ELISAs (manufactured by Ortho, Abbott, and UBI, tests II-IV) were compared. Sera from 4 different sources were used: (1) intravenous drug-users (IVDUs, n = 50), (2) blood donors (n = 1055), (3) all clinical samples from one day of routine anti-HCV testing (n = 89), (4) hemodialysis patients previously found negative by test I but clinically suspected to have a HCV infection (n = 11). Confirmatory anti-HCV tests were carried out with a second generation recombinant immunoblot assay (RIBA II). In sera positive exclusively by test IV, one antibody consumption test (UBI HCV Neutralization EIA) and one further immunoblot assay (INNO-LIA HCV Ab) were used. PCR for HCV RNA was carried out on all hemodialysis patient sera and in the RIBA II positive blood donor sera. The second generation ELISAs discriminated 11 more positive samples than the first generation test (2 IVDUs, 5 blood donors, 4 clinical samples). The 9 sera from blood donors and clinical samples were all RIBA II positive or indeterminate. The second generation tests thus showed increased sensitivity. The second generation tests also showed increased specificity in that 4 samples that were positive by test I but negative by the second generation tests, were also negative by RIBA II. With few exceptions, all RIBA II-positive and most of the indeterminate samples were positive by the second generation ELISAs. With few exceptions, all the RIBA II-negative samples were negative by the second generation ELISAs. Eleven blood donor sera were positive by test IV exclusively where RIBA II and other supplementary assays were negative. The recently introduced second generation anti-HCV ELISAs were found to have a higher sensitivity than the first generation test. The tests also showed a good concordance with the exception of test IV in the group of blood donor sera.     

Evaluation of a new screening assay for HTLV‐1 and ‐2 antibodies for large‐scale use

Laboratory testing for Human T‐lymphotropic Virus type 1 and 2 (HTLV‐1 and ‐2) infections has become routine in blood transfusion, tissue transplantation and clinical diagnoses in many countries worldwide. Screening is usually based on the detection of antibodies to HTLV‐1 and/ or ‐2. The number of commercially available assays is limited, and among them, ELISA tests based on microtiter format are most commonly used. Recently, the new rHTLV‐I/II assay (Abbott Laboratories, Abbott Park, IL) was released; this assay was developed for an automatic large‐scale screening platform. This assay was evaluated using pre‐characterized serum panels and routine samples from the clinical laboratory. The sensitivity was 100% for HTLV‐1 and ‐2 (99/99 and 42/42, respectively, including one sample that was dually reactive, HTLV‐1 + 2). To test assay specificity, panels of blood donor sera, specimens from patients with autoimmune diseases and some viral infections were used. False‐reactive samples from previous HTLV diagnoses were also included. With these panels, the specificity was 99.4% (619/623). However, the four false‐reactive samples all belonged to the group of samples that were previously considered as false‐reactive for HTLV‐antibodies. All other samples were negative by the rHTLV‐I/II assay, and thus 100% specificity was obtained. The 1,412 samples tested in the clinic by this assay in routine use were all negative (100% specificity). Taken together, the overall specificity was 99.8%. The assay was sensitive, specific and appropriate for the large‐scale screening of samples for HTLV‐1/2 antibodies. J. Med. Virol. 82:1606–1611, 2010. © 2010 Wiley‐Liss, Inc.     

Prevalence of antibodies to HTLV in antenatal clinic attenders in South East London

The prevalence of antibodies to HTLV in women attending a south east London antenatal clinic between October 1990 and July 1992 was determined using sera referred for routine rubella antibody testing. Samples were screened for HTLV antibody using a modified Fujirebio gel particle agglutination test and reactive sera confirmed by ELISA (Abbott Laboratories, North Chicago, IL) and two commercial Western blots (Cambridge Biotech Inc., Rockville, MD, and Diagnostic Biotechnology, Genelab Diagnostics, Louvaine, Belgium). This strategy confirmed the presence of HTLV-1 antibodies in 12 out of 6,289 sera (0.19%, 95% confidence limits 0.083% to 0.30%) and HTLV-2 antibodies in 2 (0.03%) sera. Specimens from 8 of 821 (0.97%, 95% confidence limits 0.42% to 1.9%) Afro-Caribbean women, three of 1,136 (0.26%, 95% confidence limits 0.055% to 0.78%) African women, and one of 3,049 (0.033%, 95% confidence limits 0.006% to 0.18%) Caucasian women were positive for HTLV-1 antibodies. Sera from Afro-Caribbean women born in the Caribbean were 7.6 times more likely to be HTLV-1 antibody positive than sera from Afro-Caribbean women born in the UK (P = 0.012). Selective testing of Afro-Caribbean and African antenatal clinic attenders, in this setting, would have identified 11 of the 12 HTLV-1 infections at an estimated cost of prevention of HTLV-1 associated disease of £100,000 per case which is considerably less than the £1.3 million which has been estimated to prevent a case by universal screening of UK blood donors. J. Med. Virol. 52:326–329, 1997. © 1997 Wiley-Liss, Inc.     

HTLV infected individuals have increased B-cell activation and proinflammatory regulatory T-cells

Human T-lymphotropic virus (HTLV) affects the human immune system in many ways, most notably by inducing proliferation of infected CD4 + T cells, but several other cell types are also affected. To characterize the effects of HTLV infection, we analysed blood samples from HTLV-infected individuals by flow cytometry. Samples were collected from visitors at the HIV clinic in Bissau, Guinea-Bissau. These samples were tested for HTLV and HIV, and 199 were analysed by flow cytometry using panels for B cells, T-cell maturation and activation, regulatory T cells (Tregs) and monocytes. CD80+ cell proportions were significantly higher in HTLV infected than in HTLV uninfected in all B cell subsets. Among T cells, there was no change in cell distribution between maturation stages, but a higher CD25+ proportion among Tregs (61.1 % vs 36.3 %, p < 0.001) in HTLV infected than in HTLV uninfected. The level of CD49d on individual cells was also higher (MFI 2734.5 vs 1,041, p < 0.001). In HTLV infected individuals, CD8 + T cells had a lower proportion of CTLA-4+ (2.5 % vs 3.5 %, 0.048) and higher PD1+ proportion on the CD45RO + subset (81.6 % vs 77.1 %, p < 0.001). Together, these findings point toward reduced regulation in HTLV + patients, which leads to immune activation. This study corroborates previous findings and offers new insight into the effects of HTLV by providing a broad flowcytometric analysis of immune cells in HTLV + individuals.     

The accuracy of an alternative confirmatory strategy for detection of antibodies to HIV-1: experience from a regional laboratory in Kagera, Tanzania.

BACKGROUND: Constant improvement of HIV tests often results in withdrawal of poorer quality tests by the manufacturing companies. It is thus often necessary to evaluate new HIV testing kits and modify the existing testing strategies. OBJECTIVES: To evaluate an alternative HIV antibody testing strategy which involves consecutive testing of sera by two enzyme-linked immunosorbent assays (ELISA), which both are recombinant antigen-based but utilise different test principles, followed by re-testing of sera giving discordant results. STUDY DESIGN: Sera (n = 1558) from a cross-sectional study of the HIV-1 seroprevalence in the Kagera region of Tanzania were tested using two ELISAs in parallel: Enzygnost anti-HIV-1/2 plus and Wellcozyme HIV-1 recombinant. Western blot analysis was done on all concordantly reactive and repeatedly discordant reactive samples as well as on 10% of concordantly ELISA negative sera. RESULTS: Two hundred and four sera (13.1%) were confirmed HIV-1-antibody positive. Both ELISAs had a sensitivity of 100%. The specificities of the ELISAs at initial and repeated testing were 99.8 and 99.9%, respectively, for Enzygnost and 97.7 and 99.5%, respectively, for Wellcozyme. None of the sera was concordantly false positive in both ELISAs. The mean ratio of the optical density of a sample to the cut off value of the test run (OD/CO ratio) was lower for samples giving false positive reactions than for confirmed HIV-1-antibody-positive samples. It is therefore important to interpret with caution HIV antibody ELISA test results on samples giving low OD/CO ratios. None of 10% of randomly selected concordantly ELISA negative sera gave a positive Western blot reaction. CONCLUSIONS: This field evaluation of an HIV antibody testing strategy involving the use of a recombinant antigen-based sandwich ELISA (Enzygnost) followed by a recombinant antigen-based competitive ELISA (Wellcozyme) showed that it had a sensitivity and specificity of 100%.     

Site-directed ELISA with synthetic peptides representing the HIV transmembrane glycoprotein

Two partially overlapping 19 and 22 amino acids long peptides representing a highly immunogenic site of the transmembranous glycoprotein (gp41) of human immunodeficiency virus (HIV) were used as antigen in ELISA tests. The results of antibody determination with this assay were compared with those of three or more conventional ELISAs and Western blot (WB) tests and radioimmunoprecipitation assay. Twenty-six sera from patients with AIDS or LAS and from asymptomatic carriers of HIV infection all showed a pronounced reaction in the peptide ELISA as well as positive results with other tests. In contrast, 27 sera from laboratory workers and blood donors were negative by all tests. A group of 39 blood donor sera, which had shown false positive or ambiguous results in the ELISAs and sometimes in WB tests employed for confirmation, also were negative in all cases with the peptide ELISA. Consecutive samples collected from individuals with primary HIV infection were also analyzed. In 6 out of 9 cases, the peptide ELISA revealed an antibody response within one month after onset of clinical symptoms and sensitivity for antibody detection equaled that of other ELISA tests. Eight sera from five West African persons infected with HIV-related viruses did not react in the peptide ELISA, reflecting differences in properties of the envelope components. The peptide ELISA used in this study appears to represent a simple technique employing chemically synthesized antigen for accurate and sensitive estimation of antibodies to the HIV group of nontransforming human retroviruses.     

Antibodies reactive with human immunodeficiency virus gag-coded antigens (gag reactive only) are a major cause of enzyme-linked immunosorbent assay reactivity in a blood donor population

Normal blood donors were examined for human immunodeficiency virus (HIV)-reactive antibodies with both virus- and Escherichia coli-expressed env- and gag-coded antigens. The frequency of samples from normal (low-risk) donors that were repeatedly reactive with an HIV enzyme-linked immunosorbent assay blood screening test (Du Pont Co.) was 0.6%. Two classes of HIV serological reactivity were identified: a minor env-reactive class (0.03 to 0.06% of donors) and the predominant env-nonreactive gag-reactive class (gag reactive only [GRO]) (0.4 to 0.5% of donors). Assignment of env reactivity was made by a synthetic (recombinant) env enzyme-linked immunosorbent assay and virus immunoblot. Most GRO sera reacted with p15/p17 bands on HIV immunoblot. Antibody specificity in GRO sera was confirmed by competition-binding studies with viral gag and E. coli-expressed p55gag. This study provides independent verification that gag-specific antibodies are present in many env-nonreactive sera. More serological and virological studies of individuals with this antibody pattern should be pursued to determine the origin of these gag-reactive antibodies.     

Quantitative detection of human immunodeficiency virus (HIV) antigen by the Enzymun-Test: comparison with alternative assays and nucleic acid sequence-based amplification of HIV type 1 RNA.

A new modular automated enzyme immunoassay (EIA) (Enzymun-Test HIV Ag: Boehringer Mannheim) for quantitative human immunodeficiency virus (HIV) antigen detection was evaluated by testing a panel of 1,506 serum samples, including seroconversions, dilution series, follow-up samples from patients under antiretroviral therapy, single serum specimens from HIV-seropositive individuals in different stages of infection, potentially cross-reactive samples, and sera from HIV-negative hospitalized patients. The Abbott HIV type 1 (HIV-1) antigen monoclonal antibody assay served as the reference assay, and nucleic acid sequence-based amplification (Organon Teknika) for quantitative amplification of HIV-1 RNA was used for follow-up of patients under antiretroviral chemotherapy. The Boehringer Mannheim and Abbott EIAs showed concordant results for the early detection of HIV antigen in all the seroconversion panels. The follow-up samples from 29 HIV-infected individuals under antiretroviral therapy gave divergent results between both antigen tests. For the detection of HIV antigen in single serum samples from HIV-infected patients in different stages of HIV infection, a higher number of positive samples was detected with the Abbott HIV-1 antigen monoclonal antibody assay in samples from patients in stages II and III of HIV infection. The Enzymun-Test detected three or more positive samples than did the Abbott assay among the samples of patients with AIDS. The concordance on a sample-to-sample basis between the Boehringer Mannheim and Abbott EIAs was 98.6%. The sensitivity of the Enzymun-Test in comparison to the reference assay was 97.2%; the specificity was 98.8%. Although no close correlation could be found between the amount of viral RNA in serum detected by nucleic acid sequence-based amplification and the concentration of HIV antigen, a high HIV-1 RNA copy number was mostly associated with high levels of HIV antigen. In conclusion, the Enzymun-Test permits accurate HIV antigen detection and offers, in contrast to previous assays, the possibility of completely automated detection.     

Performance evaluation of the Abbott HTLV III EIA, a test for antibody to HTLV III in donor blood

An enzyme immunoassay (EIA), designed to detect antibodies to human T-cell lymphotropic virus type III (HTLV III), was evaluated. The antibody test was found to be highly sensitive; serum from 221 of 223 (99.1%) patients with acquired immune deficiency syndrome (AIDS) was positive for antibodies to HTLV III. In addition, the antibody test was found to be highly specific; approximately 99.75% of 20,720 serum or plasma samples from blood donors were negative for antibody to HTLV III. In most cases, the Western Blot analysis agreed well with the EIA. Eighty-one of 82 (98.8%) EIA-positive samples from patients with AIDS were Western Blot positive. Of the EIA-positive blood donors, 21 of 36 (58%) were detected and confirmed by Western Blot analysis. A solid-phase competitive EIA also was evaluated as an alternate procedure. The preliminary results indicate that this immunoassay has a high degree of sensitivity and specificity and could serve as an alternate procedure to detect antibody to HTLV III.     

Performance of six commercial enzyme immunoassays and two alternative HIV-testing algorithms for the diagnosis of HIV-1 infection in Kisumu, Western Kenya

Performances of serological parallel and serial testing algorithms were analyzed using a combination of three ELISA and three rapid tests for the confirmation of HIV infection. Each was assessed individually for their sensitivity and specificity on a blinded panel of 769 retrospective sera of known HIV status. Western blot was used as a confirmatory assay for discordant results. Subsequently, one parallel and one serial testing algorithm were assessed on a new panel of 912 HIV-positive and negative samples. Individual evaluation of the ELISAs and rapid tests indicated a sensitivity of 100% for all assays except Uni-Gold with 99.7%. The specificities ranged from 99.1% to 99.4% for rapid assays and from 97.5% to 99.1% for ELISAs. A parallel and serial testing algorithms using Enzygnost and Vironostika, and Determine followed by Uni-Gold respectively, showed 100% sensitivity and specificity. The cost for testing 912 samples was US$4.74 and US$ 1.9 per sample in parallel and serial testing respectively. Parallel or serial testing algorithm yielded a sensitivity and specificity of 100%. This alternative algorithm is reliable and reduces the occurrence of both false negatives and positives. The serial testing algorithm was more cost effective for diagnosing HIV infections in this population.