FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1906692317> ?p ?o ?g. }

• W1906692317 endingPage "222" @default.
• W1906692317 startingPage "214" @default.
• W1906692317 abstract "Objectives: To compare the diagnostic usefulness in tuberculosis of the serodiagnostic enzyme-linked immunosorbent assay (ELISA) kit A60 (Anda Biologicals, Strasbourg, France) and of our domestic ELISA based on three purified cell wall glycolipid antigens.Methods: The presence and concentrations of IgG and IgM anti-A60 antibodies and anti-LOS, anti-DAT and anti-PGLTb1 antibodies against the glycolipid antigens were determined by ELISA in 50 HIV-seronegative and 46 HIV-seropositive patients, with documented active tuberculosis. The specificity of these ELISAs was determined with use of sera from 50 healthy blood donors, 29 patients with non-mycobacterial pulmonary diseases and 24 HIV-positive patients with disseminated Mycobacterium avium infection.Results: With a calculated cut-off for each antigen and immunoglobulin that gave a specificity higher than or equal to 98%, the cumulative ELISA results showed that only 36.5% of the patients with tuberculosis had a positive response in the A60 test, as compared with 84.4% who showed a response to the three glycolipid antigens (p<0.001). This striking difference persisted when the cumulative sensitivities were calculated according to the HIV status of the patients and the localization of tuberculosis. The anti-A60 antibody (IgG and IgM) levels and the degree of sensitivity of the ELISA for detection of A60 antigen were always lower in HIV-positive patients with pulmonary and extrapulmonary tuberculosis than in HIV-negative patients with tuberculosis. The sensitivity of A60 ELISA was further decreased in HIV-positive patients with low CD4+ lymphocytes counts, in contrast to the results with the three glycolipid antigens.Conclusions: These results show the limitations of the A60 ELISA, and confirm the potencies of the glycolipid antigens in serodiagnosis of tuberculosis in HIV-positive and HIV-negative patients. Objectives: To compare the diagnostic usefulness in tuberculosis of the serodiagnostic enzyme-linked immunosorbent assay (ELISA) kit A60 (Anda Biologicals, Strasbourg, France) and of our domestic ELISA based on three purified cell wall glycolipid antigens. Methods: The presence and concentrations of IgG and IgM anti-A60 antibodies and anti-LOS, anti-DAT and anti-PGLTb1 antibodies against the glycolipid antigens were determined by ELISA in 50 HIV-seronegative and 46 HIV-seropositive patients, with documented active tuberculosis. The specificity of these ELISAs was determined with use of sera from 50 healthy blood donors, 29 patients with non-mycobacterial pulmonary diseases and 24 HIV-positive patients with disseminated Mycobacterium avium infection. Results: With a calculated cut-off for each antigen and immunoglobulin that gave a specificity higher than or equal to 98%, the cumulative ELISA results showed that only 36.5% of the patients with tuberculosis had a positive response in the A60 test, as compared with 84.4% who showed a response to the three glycolipid antigens (p<0.001). This striking difference persisted when the cumulative sensitivities were calculated according to the HIV status of the patients and the localization of tuberculosis. The anti-A60 antibody (IgG and IgM) levels and the degree of sensitivity of the ELISA for detection of A60 antigen were always lower in HIV-positive patients with pulmonary and extrapulmonary tuberculosis than in HIV-negative patients with tuberculosis. The sensitivity of A60 ELISA was further decreased in HIV-positive patients with low CD4+ lymphocytes counts, in contrast to the results with the three glycolipid antigens. Conclusions: These results show the limitations of the A60 ELISA, and confirm the potencies of the glycolipid antigens in serodiagnosis of tuberculosis in HIV-positive and HIV-negative patients. Human tuberculosis remains a common disease in developing countries. In Europe and the USA its incidence had decreased during the last decade, but is now rising following the emergence of AIDS [1Festenstein F Grange JM Tuberculosis and the acquired immune deficiency syndrome.J Appl Bacteriol. 1991; 71: 19-30Crossref PubMed Google Scholar], the deterioration of control measures and the increase in untested immigrants [2McKenna MT McCray E Onorato I The epidemiology of tuberculosis among foreign-born persons in the United States, 1986 to 1993.N Engl J Med. 1995; 332: 1071-1076Crossref PubMed Scopus (351) Google Scholar]. It is estimated that more than three million people worldwide are infected with both Mycobacterium tuberculosis and HIV [3Raviglione MC Snider DE Kochi A Global epidemiology of tuberculosis. Morbidity and mortality of a worldwide epidemic.JAMA. 1995; 273: 220-226Crossref PubMed Scopus (1506) Google Scholar]. Among the diseases associated with HIV infection, tuberculosis (TB) is of particular importance. It is often the first clinical manifestation of immunodeficiency, it is highly transmissible and it is associated with rapid deterioration of the immune competence of the host. However TB is potentially preventable by chemoprophylaxis and is readily treatable [4Pape JW Jean SS Ho JL Hafner A Johnson WD Effect of isoniazid prophylaxis on incidence of active tuberculosis and progression of HIV infection.Lancet. 1993; 342: 268-272Abstract PubMed Scopus (421) Google Scholar]. Screening for TB at HIV testing sites could be an effective approach to early detection of active TB not just among HIV-positive, but also in HIV-negative, individuals [5Espinal MA Reingold AL Koenig E Lavandera M Sanchez S Screening for active tuberculosis in HIV testing centre.Lancet. 1995; 345: 890-893Abstract PubMed Google Scholar]. However, the discriminative values of the tuberculin skin test vary with the population tested, and furthermore the reaction to purified protein derivate cannot be read before 48 to 72 h and the required return visit would make the process of TB screening cumbersome. As an alternative, several immunologic methods have been used to detect serum antibodies against mycobacterial antigens in TB patients [6Grange JM The humoral immune response in tuberculosis: its nature, biological role and diagnostic usefulness.Adv Tuberc Res. 1984; 21: 1-78PubMed Google Scholar]. The enzyme-linked immunosorbent assay (ELISA) has been widely used for the diagnosis of active TB [7Daniel TM Debane SM The serodiagnosis of tuberculosis and other mycobacterial diseases by enzyme-linked immunosorbent assay.Am Rev Resp Dis. 1987; 135: 1137-1151PubMed Google Scholar], and can evaluate the response in different immunoglobulin classes, IgG being the most discriminatory when TB patients are compared with healthy control subjects [8Grange JM Gibson J Nassau E Kardjito T Enzyme-linked immunosorbent assay (ELISA): a study of antibodies to M. tuberculosis in the IgG, IgA and IgM classes in tuberculosis, sarcoidosis and Crohn's disease..Tubercle. 1980; 61: 145-152Abstract Full Text PDF PubMed Scopus (73) Google Scholar, 9Daleine G Lagrange PH Preliminary evaluation of M. tuberculosis lipooligosaccharide (LOS) antigen in the serological diagnosis of tuberculosis in HIV seropositive and seronegative patients..Tubercle Lung Dis. 1995; 76: 234-239Abstract Full Text PDF PubMed Scopus (19) Google Scholar]. During the last 15 years several purified antigens from M. tuberculosis have been used for the serodiagnosis of tuberculosis [10Wilkins EGL The serodiagnosis of tuberculosis.in: Davies PDO Clinical tuberculosis. Chapman and Hall, London1994: 367-380Google Scholar]. They include a high-molecular-weight mycobacterial antigen complex, known as A60 [11Cocito C Vanlinden E Preparation and properties of antigen 60 from Mycobacterium bovis BCG.Clin Exp Immunol. 1986; 66: 262-272PubMed Google Scholar]. The detection of immunoglobulin G (IgG) antibodies to A60 has been advocated as a reasonably sensitive and specific test for active tuberculosis [12Baelden MC Vanderelst B Dieng M Prignot J Cocito C Serological analysis of human tuberculosis by an ELISA with mycobacterial antigen 60.Scand J Infect Dis. 1990; 22: 63-73Crossref PubMed Scopus (39) Google Scholar]. It is the only commercially available ELISA test for the diagnosis of tuberculosis in France. However, there are contradictory reports about the usefulness of A60 ELISA in HIV-positive patients [13VanDerWerf TS Das PK VanSooligen D et al.Serodiagnosis of tuberculosis with A60 antigen enzyme-linked immunosorbent assay: failure in HIV-infected individuals in Ghana.Med Microbiol Immunol. 1992; 181: 71-76Crossref PubMed Scopus (18) Google Scholar]. The aim of this study was to compare the discriminative values of an ELISA test using A60 with those obtained with a domestic ELISA test, which has already been described [9Daleine G Lagrange PH Preliminary evaluation of M. tuberculosis lipooligosaccharide (LOS) antigen in the serological diagnosis of tuberculosis in HIV seropositive and seronegative patients..Tubercle Lung Dis. 1995; 76: 234-239Abstract Full Text PDF PubMed Scopus (19) Google Scholar, 14Simonney N Molina JM Molimard M Okesenhendler E Perronne C Lagrange PH Analysis of the immunological humoral response to M. tuberculosis glycolipid antigens (DAT, PGLTbl) for diagnosis of tuberculosis in HIV seropositive and seronegative patients.Eur J Clin Microbiol Infect Dis. 1995; 14: 883-891Crossref PubMed Scopus (35) Google Scholar], using three glycolipid antigens specific for M. tuberculosis. These were applied to the same serum specimens taken from HIV-negative and HIV-positive patients with documented (culture-positive) tuberculosis before they received any treatment. Between May 1991 and May 1994, acid-fast bacteria were isolated from 120 hospitalized patients (aged 21 to 85 years) and identified by procedures recommended by reference laboratories. Of these patients, 96 were infected with M. tuberculosis and 24 with mycobacteria other than tuberculosis (MOTT), mostly M. aviumintracellulare. Among the 96 patients with documented tuberculosis, 22 were female and 74 were male. Forty-six patients with tuberculosis had a pulmonary localization (21 HIV positive and 25 HIV negative), and 50 had extrapulmonary tuberculosis (25 HIV positive and 25 HIV negative). Among the 25 HIV-negative patients with extrapulmonary disease, the localizations of tuberculosis were lymphatic tissues (12), pleural effusion (4), bone (3), gastrointestinal tract (2), disseminated tuberculosis (2), central nervous system (1) and skin (1). Among the 25 HIV-positive patients, 15 patients had disseminated tuberculosis, 9 had lymph node involvement, and one had a pleural effusion. The mean ages were similar in the different subgroups (39.2 ± 2.2 versus 36.6 ± 1.9 in HIV-positive patients with extrapulmonary tuberculosis and pulmonary tuberculosis, and 43.9 ± 3.3 versus 40.9 ± 2.5 in HIV-negative patients with extrapulmonary and pulmonary tuberculosis, respectively). CD4+ T lymphocyte counts were known in 38 HIV-positive patients within 3 months of the time of diagnosis of tuberculosis and there was no statistically significant difference for the mean ± SEM of CD4+ lymphocytes between patients with pulmonary (89.9 ± 22.9) and patients with extrapulmonary TB (88.3 ± 20.5). The median CD4+ T-lymphocyte counts were also equivalent (50 CD4+/mm3 and 47 CD4+/mm3, respectively). All the MOTT-infected patients were HIV positive and presented disseminated atypical mycobacterial diseases. Sera from 50 healthy blood donors and from 29 patients with pulmonary diseases but without documented mycobacterial isolation (12 being HIV negative and 17 being HIV positive) were used as normal and pathologic controls, respectively. All sera were collected from patients in the TB-HIV−-, TB-HIV+ and other mycobacteriosis groups, on the day when specimens were obtained for the bacteriologic diagnosis of mycobacteriosis. All sera were conserved at −20°C until tested. The isolation, purification and characterization of the LOS (fractions 4 and 5 of polar glycolipids belonging to the lipo-oligosaccharide family), DAT (2,3-diacyltrehalose, formerly referred to as SL IV antigen), and PGLTb1 (triglycosyl phenol phthiocerol dimycocerosate) were carried out as previously reported [15Daffe M Lacave C Lanéelle MA Lanéelle G Structure of the major triglycosyl phenol-phtiocerol of M. tuberculosis (strain Canetti).Eur J Biochem. 1987; 167: 155-160Crossref PubMed Scopus (116) Google Scholar, 16Daffé M Papa A David HL Glycolipids of recent clinical isolates of M. tuberculosis: chemical characterisation and immunoreactivity.J Gen Microbiol. 1989; 135: 2759-2766PubMed Google Scholar, 17Lemassu A Lanéelle MA Daffe M Revised structure of a trehalose-containing immunoreactive glycolipid of M. tuberculosis.FEMS Microbiol Lett. 1991; 78: 171-176Crossref Scopus (45) Google Scholar, 18Papa F Cruaud P Luquin M et al.Isolation and characterization of serology reactive lipooligosaccharides from M. tuberculosis.Res Microbiol. 1993; 144: 91-99Crossref PubMed Scopus (13) Google Scholar]. The three glycolipid antigens used in our ELISA test were kindly supplied by Dr F. Papa (Unité de Tuberculose et Mycobacteries, Institut Pasteur, Paris, France). A60 is a commercialy available enzyme-linked immunosorbent assay (ELISA, Anda Biological, Strasbourg, France). The glycolipid ELISA test was performed in polystyrene microtiter plates from Costar (France), by a previously described method with some modifications that have already been published [14Simonney N Molina JM Molimard M Okesenhendler E Perronne C Lagrange PH Analysis of the immunological humoral response to M. tuberculosis glycolipid antigens (DAT, PGLTbl) for diagnosis of tuberculosis in HIV seropositive and seronegative patients.Eur J Clin Microbiol Infect Dis. 1995; 14: 883-891Crossref PubMed Scopus (35) Google Scholar]. At the same time, IgM and IgG against A60 were determined in an ELISA test, according to the manufacturer's instructions. Final calculations were performed as already described by Cruaud et al. [19Cruaud P Yamashita JT Martin-Casabona N et al.Evaluation of a novel 2,3-diacyl-trehalose-2′ sulfate (SL-IV) antigen for case finding and diagnosis of leprosy and tuberculosis.Res Microbiol. 1990; 141: 679-694Crossref PubMed Scopus (45) Google Scholar]. Arithmetic mean, median and standard error of the mean (SEM) were calculated for optical density values. Statistical significance was determined using the Quattro Pro 5.0 program (Borland, France). The significance of differences between continuous, normally distributed variables was calculated using the Student's t-test. Sensitivity, specificity and predictive values were estimated as recommended [20Vecchio TJ Predictive value of a single diagnostic test in unselected populations.N Engl J Med. 1966; 274: 1171-1173Crossref PubMed Scopus (705) Google Scholar, 21McNeil BJ Keeler E Adelstein SJ Primer on certain elements of medical decision-making.N Engl J Med. 1975; 293: 211-215Crossref PubMed Scopus (882) Google Scholar]. Frequencies were compared using the chi-squared test with Yates and Bonferroni corrections [22Koopmans LH Introduction to contemporary statistical methods. Duxbury Press, Boston1987Google Scholar]. Cut-off points for the sensitivity and specificity of the ELISA with each antigen and immunoglobulin were calculated from the results obtained with sera from healthy blood donors and the pathologic control groups (including sera from HIV patients with disseminated MOTT disease). Among the 96 documented tuberculosis patients, 81 were positive for at least one antigen in ELISA with LOS, DAT or PGLTb1 (IgG antibody determination) in comparison with only 35 that were positive in A60 antigen tests when IgG and IgM antibody estimations were combined. The individual levels of IgG antibodies to LOS, DAT and PGLTbl antigens and of IgG and IgM antibodies to A60 antigens in patients with documented tuberculosis, in healthy people and in pathologic controls are shown in Figures 1 and 2. In general, antibody titers were higher in tuberculosis patients than in healthy and pathologic controls (p<0.001). There were very few false positive reactions among healthy donors and pathologic controls. However, approximately one-third of false-negative results were obtained with the three glycolipid antigen ELISA, and rather more than half when A60 ELISA was used, in patients with tuberculosis.Figure 2Individual serum levels of IgM (A) and IgG (B) antibody to A60 antigen from the same groups as indicated in Figure 1.View Large Image Figure ViewerDownload (PPT) The means of specific antibody levels to each antigen (LOS, DAT, PGLTbl and A60) were calculated in subgroups of HIV-negative and HIV-positive patients with tuberculosis, according to the localization of disease and to the smear test results. The mean values and standard errors of the mean of optical densities (ODs) obtained with the sera from 46 HIV-positive and 50 HIV-negative TB patients are shown in Table 1. In general, there was no statistically significant difference between the antibody levels in multibacillary (smear positive) TB and in paucibacillary (smear-negative) TB except for the PGLTbl and A60 antigens in HIV-negative TB patients. These HIV-negative patients with multibacillary TB had significantly lower levels (p< 0.05) of anti-PGLTb1 antibody than HIV-negative patients with paucibacillary TB. The reverse was observed when levels of anti-A60 (IgG or IgM) antibodies were considered; however, only the HIV-negative patients with multibacillary TB had significantly higher levels of IgM anti-A60 antibody as compared with the HIV-negative patients with paucibacillary TB.Table 1Anti-LOS, anti-DAT, anti-PGLTb1 IgG and anti-A60 IgG and IgM antibody levels according to smear positivity or negativity in HIV-negative and HIV-positive patients with pulmonary and extrapulmonary tuberculosisOptical density (mean ± SEM)aAbsorbance at 414 nm × 10−3; ±SEM (standard error of the mean).HIV-negative patientsHIV-positive patientsAntigenSmear testTotalPulmonaryExtrapulmonaryTotalPulmonaryExtrapulmonaryLOS+162.0±45.9184.2±55.591.0±74.9152.5±33.5159.8±56.9147.5±42.7-103.3±15.796.3±26.8104.2±19.6171.7±22.6167.0±28.9176.4±36.1DAT+270.2±51.6312.7±64.2134.4±20.7217.4±27.5269.7±50.4181.2±28.2-203.7±27.9156.1±47.6225.2±33.9280.8±30.8250.6±30.7311.0±53.4PGLTb1+77.1 ±17.9 (P<0.001)88.3±22.741.4±14.4 (P<0.001)244.53±8.7299.3±73.4206.5±41.6-164.6±34.0125.2±39.3182.6±46.3264.2±34.4237.3+50.9291.1±58.5A60.G+533.7±136.8535.6±151.3527.6±344.1355.5±94.6255.3±91.4424.9±147.3-391.4±75.9461.1±175.9411.0±115.6231.6±69.7290.3±136.1172.9±34.5A60.M+727.9±110.6 (p<0.05)726.6±116.9729.0+303.4 (p<0.01)326.8±89.8341.1±86.9316.9±142.8-393.6±67.9468.0±134.2360.1±79.2286.9±43.2336.4±76.5237.4±39.0a Absorbance at 414 nm × 10−3; ±SEM (standard error of the mean). Open table in a new tab The sensitivities of the ELISAs using the IgG assays for the three glycolipid antigens were compared with the sensitivities of the ELISA using the IgM and IgG assays for the A60 antigen in the 96 patients with documented tuberculosis, independently of their HIV status (Table 2). The results showed that A60 ELISA for IgG or for IgM antibodies was always less sensitive (28.1% and 17.7%, respectively) than the glycolipid ELISA for IgG (anti-LOS, 64.6%; anti-DAT, 61.5%; anti-PGLTbl, 54.2%). A sample was considered positive if a positive response to at least one of the three glycolipid antigens or to at least one of the IgG and IgM determination for A60 was detected. The cumulative sensitivity of two combined tests was always higher (p<0.01) when two glycolipid antigens were used, than when IgG and IgM A60 tests were combined (Table 2). The higher cumulative sensitivity was due to the fact that some patient sera reacted with only one antigen, while some react with all three glycolipid antigens. The cumulative sensitivity was further increased when the sensitivity was calculated using the combined results of the three glycolipid antigens, reaching 84.4% (p<0.05).Table 2Sensitivity of the ELISA for detection of anti-A60 IgG or IgM and of anti-LOS, anti-DAT and anti-PGLTb1 IgG antibodies in various study groupsActive tuberculosis (n=96)Active tuberculosis HIV-negative (n=50)Active tuberculosis HIV-positive (n=46)AntigenNo. positivePercent sensitivityNo. positivePercent sensitivityNo. positivePercent sensitivityA60 (IgG)2728.11836.0919.6A60 (IgM)1717.71224.0510.9LOS (IgG)6264.62856.03473.9DAT (IgG)5961.52754.03269.6PGLTb1 (IgG)5254.21632.03678.3A60 (IgG + IgM)3536.52346.01123.9DAT + LOS (IgG)7780.23774.04086.9PGLTb1 + LOS (IgG)7477.13468.04086.9DAT + PGLTb1 (IgG)6971.93162.23882.6DAT + PGLTb1 + LOS (IgG)8184.43978.04291.3Cut-off points: LOS, 65; DAT, 150; PGLTb1, 126; A60.G, 360; A60.M, 684. Open table in a new tab Cut-off points: LOS, 65; DAT, 150; PGLTb1, 126; A60.G, 360; A60.M, 684. Only one serum sample from a healthy blood donor was positive for DAT and PGLTb1. The specificities calculated from the results in healthy controls were 98% for DAT and PGLTb1 and 100% for LOS and A60 IgG and IgM. Such high specificity (98%) remained when the results of the three antigens (LOS, DAT, PGLTb1) or those of IgM and IgG ELISA A60 were combined (Table 3). When ELISA for the three glycolipid antigens was considered, high specificity (100%) was observed in pulmonary disease HIV-negative controls, but it dropped to 82.4% in the HIV-positive pulmonary disease controls and was further diminished to 62.5% in HIV-positive patients with disseminated atypical mycobacteriosis. Such reduced specificity was not observed when ELISA for A60 antigen was performed in these pathologic sera but was about 80% in HIV-positive or HIV-negative patients with localized or disseminated MOTT infection.Table 3Specificity of the ELISA for detection of A60 (IgG + IgM) and three glycolipid antigens (LOS + DAT + PGLTb1-IgG) in the various control groupsA60 (IgG + IgM)LOS + DAT + PGLTb1 (IgG)Study groupNo of sera testedNo. positivePercent specificityNo. positivePercent specificityHIV negativePulmonary disease12375.00100.0Blood donors500100.0198.0HIV positivePulmonary disease17382.4382.4Atypical mycobacteriosis24483.3962.5Cut-off points: A60G, 360; A60M, 684; LOS, 65; DAT, 150; PGLTb1, 126. Open table in a new tab Cut-off points: A60G, 360; A60M, 684; LOS, 65; DAT, 150; PGLTb1, 126. HIV-positive patients with tuberculosis had a statistically significantly greater response (p<0.05) to each of the glycolipid antigens in general (73.9% for LOS, 69.6% for DAT and 78.3% for PGLTbl) than HIV-negative patients with tuberculosis (56%, 54% and 32% for LOS, DAT and PGLTb1, respectively)(Table 2). The opposite was observed when A60 antigen was used: the positive responses were always more numerous in HIV-negative patients with tuberculosis than in HIV-positive patients with tuberculosis. The A60 IgG determinations gave almost twice as many positive results as the A60 IgM determinations, but only one third of the HIV-positive patients responded to this antigen. When combined ELISA test results were analyzed, similar results were obtained. Among the 96 TB patients, 46 had pulmonary and 50 extrapulmonary tuberculosis; 43 (44.8%) were smear positive and 53 (55.2%) were smear negative. In the group of 50 HIV-negative tuberculosis patients, 64% of pulmonary tuberculosis cases were smear-positive, compared to only 20% of extrapulmonary cases (p<0.01), giving an overall sensitivity of the smear test of 42% for the HIV-negative patients. In contrast, there was no statistically significant difference for smear test sensitivity between pulmonary (42.9%) and extrapulmonary (52.0%) tuberculosis cases in the HIV-positive group of patients. In HIV-positive patients with active tuberculosis the ELISA sensitivities to the three glycolipid antigens were statistically significantly higher than the ELISA sensitivities to A60 and were influenced neither by the localization of tuberculosis nor by the smear test results. In the HIV-negative patients with tuberculosis, the differences were less pronounced but still statistically significant when tuberculosis localization was compared, but not when the smear test results were considered (Table 4).Table 4Sensitivity of the ELISA for IgG anti-glycolipid (LOS, DAT, PGLTb1) antibodies and of IgG + IgM antibodies to A60 in the various study groups according to HIV serostatus, localization of disease and smear test results in patients with documented active tuberculosisELISANo. of patients giving positive results (%)HIV-positive patients (n=46)HIV-negative patients (n=50)LOS+DAT+PGLTb142 (91.3)39 (78.0)p<0.0001p<0.0001A60 (IgG + IgM)11 (23.9)23 (46.0)Pulmonary (n=21)Extrapulmonary (n=25)Pulmonary (n=25)Extrapulmonary (n=25)LOS+DAT+PGLTbl20 (95.2)22 (88)20 (80)19 (76)p<0.0001p<0.0001p<0.0003p<0.005A60 (IgG + IgM)6 (28.6)5(20)11 (44)12 (48)Smear + (n=9)Smear - (n=12)Smear + (n=13)Smear - (n=12)Smear + (n=16)Smear - (n=9)Smear + (n=5)Smear - (n=20)LOS + DAT + PGLTbl9 (100)11 (91.6)11 (84.6)11 (91.6)14 (87.5)6 (66.7)3 (60.0)16 (80.0)A60 (IgG + IgM)2 (22.2)4 (33.3)4 (30.8)1 (8.3)8 (50.0)3 (33.3)4 (80.0)8 (40.0)Cut-off points: LOS, 65; DAT, 150; PGLTb1, 126; A60 IgG, 360; A60 IgM, 684. Open table in a new tab Cut-off points: LOS, 65; DAT, 150; PGLTb1, 126; A60 IgG, 360; A60 IgM, 684. For all antigens except the LOS, the sensitivity of the ELISA was significantly higher in patients who had 100 or more CD4+ lymphocytes/mm3 (Table 5); there were more positive ELISA test results in patients who had less than 100 CD4 lymphocytes/mm3. When the results were stratified according to the localization of tuberculosis, the differences were only statistically significant in HIV-positive patients with extrapulmonary tuberculosis and not in patients with pulmonary tuberculosis.Table 5ELISA test results in 38 HIV-positive patients with documented tuberculosis according to their circulating CD4+ lymphocytesNumber of positive/negative resultsNumber of CD4+ lymphocytes/mm3Antigens>100 (n=13)<100 (n=25)LOS8/518/72DAT11/215/10PGLTbl12/116/9A60.IgG2/112/23A60.IgM3/101/24Cut-off points: LOS, 65; DAT, 150; PGLTbl, 126; A60 IgG, 360; A60 IgM, 684. Open table in a new tab Cut-off points: LOS, 65; DAT, 150; PGLTbl, 126; A60 IgG, 360; A60 IgM, 684. The first important finding of this study was that the sensitivity of ELISA for anti-A60 IgG was statistically significantly lower (p<0.01) than the sensitivity of ELISA for IgG against LOS, DAT and PGLTb1 in the 96 patients with tuberculosis; 28.1% compared with 64.6%, 61.5% and 54.2%, respectively, gave positive test results. The sensitivity of testing for anti-A60 IgM antibodies was even lower (17.7%) than the sensitivity obtained with the anti-A60 IgG antibody. The different combinations of the ELISA results obtained with each of the three glycolipid antigens and with the A60 IgG and IgM ELISA results were analyzed in order to increase the sensitivity while keeping the specificity high. The sensitivity of the ELISA increased when the results with LOS, DAT or PGLTb1 were combined in pairs, and addition of test results obtained using the three glycolipid antigens further increased the sensitivity to 84.4%. The ELISA sensitivity in these patients increased in the same way when the results with anti-A60 IgG and anti-A60 IgM were combined, but the combined sensitivity reached only 36.5%. On the other hand, there was no increase in the false positive rate among the control subjects (Table 3). Similar findings have already been reported for combining the results obtained with three glycolipids [23Ridell M Wallersholm G Millikin DE et al.A comparative serological study of antigenic glycolipids from M. tuberculosis.Tubercle Lung Dis. 1992; 73: 101-105Abstract Full Text PDF PubMed Scopus (34) Google Scholar] or with glycolipids and antigen 5 [24Chan SL Reggiardo Z Daniel TM Girling DJ Mitchison DA Serodiagnosis of tuberculosis using an ELISA with antigen 5 and a hemagglutination assay with glycolipid antigens.Am Rev Resp Dis. 1990; 136: 385-390Crossref Scopus (34) Google Scholar]. Analysis of these results according to patient features confirmed the results of several recent studies [13VanDerWerf TS Das PK VanSooligen D et al.Serodiagnosis of tuberculosis with A60 antigen enzyme-linked immunosorbent assay: failure in HIV-infected individuals in Ghana.Med Microbiol Immunol. 1992; 181: 71-76Crossref PubMed Scopus (18) Google Scholar, 25Saltini C Massimo A Girardi E et al.Early abnormalities of the antibody response against M. tuberculosis in human immunodeficiency virus infection.J Infect Dis. 1993; 168: 1409-1414Crossref PubMed Scopus (14) Google Scholar, 26Pouthier F Perriens JH Mukadi Y et al.Anti-A60 immunoglobulin G in the serodiagnosis of tuberculosis in HIV-seropositive and seronegative patients.AIDS. 1994; 8: 1277-1280Crossref PubMed Scopus (10) Google Scholar]. The sensitivity of testing for anti-A60 (IgG, IgM) antibodies was statistically significantly lower (p<0.0033) among HIV-positive than among HIV-negative patients with tuberculosis: 23.9% compared with 46% had positive ELISA results, respectively. Such differences persisted also when TB patients were stratified for localization of tuberculosis (Table 1). The false-negative results of the A60 ELISA in HIV-positive patients were directly related to the low numbers of circulating CD4+ lymphocytes (Table 5), as already reported by Saltini et al. [25Saltini C Massimo A Girardi E et al.Early abnormalities of the antibody response against M. tuberculosis in human immunodeficiency virus infection.J Infect Dis. 1993; 168: 1409-1414Crossref PubMed Scopus (14) Google Scholar]. However, the 46% estimate of the A60 ELISA sensitivity among HIV-negative patients reported here is lower than the sensitivities previously reported for HIV-uninfected patients with active TB [12Baelden MC Vanderelst B Dieng M Prignot J Cocito C Serological analysis of human tuberculosis by an ELISA with mycobacterial antigen 60.Scand J Infect Dis. 1990; 22: 63-73Crossref PubMed Scopus (39) Google Scholar,27Gevaudan MJ Bollet C Charpin D Mallet MN Serological response of tuberculosis patients to antigen A60 of BCG.Eur J Epidemiol. 1992; 8: 666-676Crossref PubMed Scopus (40) Google Scholar, 28Charpin H Herbault H Gevaudan MJ et al.Value of ELISA using A60 antigen in the diagnosis of active pulmonary tuberculosis.Am Rev Resp Dis. 1990; 142: 380-384Crossref PubMed Scopus (62) Google Scholar, 29Amicosante M Paone G Ameglio F et al.Antibody repertoire against the A60 antigen complex during the course of pulmonary tuberculosis.Eur Respir J. 1993; 6: 816-822PubMed Google Scholar, 30Gupta S Kumari S Banwalikas JN Gupta SK Diagnostic utility of the estimation of mycobacterial antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis.Tubercle Lung Dis. 1995; 76: 418-424Abstract Full Text PDF PubMed Scopus (51) Google Scholar]. Among HIV-uninfected patients with non-TB pulmonary conditions, the detection of IgG and IgM against A60 antigen complex has been reported to be 71-94% specific for active TB [12Baelden MC Vanderelst B Dieng M Prignot J Cocito C Serological analysis of human tuberculosis by an ELISA with mycobacterial antigen 60.Scand J Infect Dis. 1990; 22: 63-73Crossref PubMed Scopus (39) Google Scholar, 27Gevaudan MJ Bollet C Charpin D Mallet MN Serological response of tuberculosis patients to antigen A60 of BCG.Eur J Epidemiol. 1992; 8: 666-676Crossref PubMed Scopus (40) Google Scholar, 28Charpin H Herbault H Gevaudan MJ et al.Value of ELISA using A60 antigen in the diagnosis of active pulmonary tuberculosis.Am Rev Resp Dis. 1990; 142: 380-384Crossref PubMed Scopus (62) Google Scholar, 30Gupta S Kumari S Banwalikas JN Gupta SK Diagnostic utility of the estimation of mycobacterial antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis.Tubercle Lung Dis. 1995; 76: 418-424Abstract Full Text PDF PubMed Scopus (51) Google Scholar]. In this report, in order to adjust the two ELISA tests for comparable specificity using the results from healthy donors, higher cut-off values were needed, which consequently reduced sensitivity. In contrast, and as already described by Martin-Casabona et al. [31Martin-Casabona N Gonzalez FT Papa F et al.Time course of anti-SLIV immunoglobulin G antibodies in patients with tuberculosis and tuberculosis-associated AIDS.J Clin Microbiol. 1992; 30: 1089-1093PubMed Google Scholar], and in a previous report [9Daleine G Lagrange PH Preliminary evaluation of M. tuberculosis lipooligosaccharide (LOS) antigen in the serological diagnosis of tuberculosis in HIV seropositive and seronegative patients..Tubercle Lung Dis. 1995; 76: 234-239Abstract Full Text PDF PubMed Scopus (19) Google Scholar], the sensitivity of testing for anti-LOS, anti-DAT and anti-PGLTb1 IgG antibodies was statistically higher (p<0.0304) among HIV-positive than among HIV-negative patients with tuberculosis: 91.3% and 78%, respectively, had positive test results. No statistically significant differences were found in the antibody levels to the three glycolipid antigens in relation to the localization of TB (Table 1). Moreover, no difference was found in the combined ELISA results in relation to the direct microscopy results in HIV-positive or HIV-negative patients with pulmonary or extrapulmonary tuberculosis. A serodiagnostic test with a high specificity would justify the start of chemotherapy in suspected tuberculosis before culture results were available. The prevalence of TB in hospital populations suspected of having TB has been reported to be 30–52% [28Charpin H Herbault H Gevaudan MJ et al.Value of ELISA using A60 antigen in the diagnosis of active pulmonary tuberculosis.Am Rev Resp Dis. 1990; 142: 380-384Crossref PubMed Scopus (62) Google Scholar, 29Amicosante M Paone G Ameglio F et al.Antibody repertoire against the A60 antigen complex during the course of pulmonary tuberculosis.Eur Respir J. 1993; 6: 816-822PubMed Google Scholar, 30Gupta S Kumari S Banwalikas JN Gupta SK Diagnostic utility of the estimation of mycobacterial antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis.Tubercle Lung Dis. 1995; 76: 418-424Abstract Full Text PDF PubMed Scopus (51) Google Scholar, 31Martin-Casabona N Gonzalez FT Papa F et al.Time course of anti-SLIV immunoglobulin G antibodies in patients with tuberculosis and tuberculosis-associated AIDS.J Clin Microbiol. 1992; 30: 1089-1093PubMed Google Scholar, 32Daniel TM de Murillo GL Sawyer JA et al.Field evaluation of enzyme-linked immunosorbent assay for the serodiagnosis of tuberculosis.Am Rev Resp Dis. 1986; 134: 662-665PubMed Google Scholar, 33Wilkins EGL Ivanyi J Potential value of diagnosis of extrapulmonary tuberculosis.Lancet. 1990; 11: 641-644Abstract Scopus (58) Google Scholar, 34Daniel TM Debanne SM Vanderkuyp F Enzyme-linked immunosorbent assay using M. tuberculosis antigen 5 and PPD for the serodiagnosis of tuberculosis..Chest. 1985; 88: 388-392Crossref PubMed Scopus (35) Google Scholar]. However, as already reported by Daniel et al. [34Daniel TM Debanne SM Vanderkuyp F Enzyme-linked immunosorbent assay using M. tuberculosis antigen 5 and PPD for the serodiagnosis of tuberculosis..Chest. 1985; 88: 388-392Crossref PubMed Scopus (35) Google Scholar], in a group of hospitalized subjects requiring investigation for tuberculosis in a university hospital such as ours the prevalence of tuberculosis would probably be closer to 15%. Assuming a prevalence of 15%, the positive and negative predictive values of the LOS, DAT and PGLTb1 antigens, using the cut-off values determined with the results obtained with the blood donors, were 88.2%, and 97.3%, respectively, when the HIV status was not considered. When the cut-off values were obtained in the HIV-negative pathologic controls, the predictive values were higher (PPV=100%; NPV= 96.3%). But the positive predictive value dropped to 47.9%, for a negative predictive value of 98.2%, if cutoff values were obtained in the HIV-positive controls with non-TB pulmonary conditions. The HIV-positive control group included patients with disseminated mycobacterial infections other than TB. Knowing that the specific glycolipid antigens did not cross-react with antisera induced by M. avium infection [35Papa F Cruaud P David HL Antigenicity and specificity of selected glycolipid fractions from M. tuberculosis.Res Microbiol. 1989; 140: 569-578Crossref PubMed Scopus (38) Google Scholar] and since antibody titers were also lower in such HIV-positive controls than in active TB patients, it is quite possible that such low persisting antibodies in HIV-positive controls reflect either non-active TB which had been treated several months before these episodes, or concomitant infection with M. tuberculosis and M. avium [36Piersimoni C Felici L Giorgi P Morbiducci V Scalise G Mixed mycobacterial infection of cervical lymph nodes.Pediatr Infect Dis J. 1991; 10: 544-545Crossref PubMed Scopus (6) Google Scholar]. This possibility is being investigated. The positive predictive values of the A60 ELISA (from 18.4% to 24.6%) become too low to be clinically useful in HIV-positive or in HIV-negative patients suspected to have tuberculosis. In contrast, the optimal sensitivity and specificity were obtained by combining the results of the ELISA with the LOS, DAT and PGLTbl antigens. Such an ELISA assay may be clinically useful in patients clinically suspected to have TB. The authors wish to thank Alexandra Dumontier for her excellent secretarial help. This work was funded in part by a research grant from ANRS (RIN 102) and in part by a research grant from the DRC (Direction de la Recherche Clinique) of the Assistance Publique, Hôpitaux de Paris." @default.
• W1906692317 created "2016-06-24" @default.
• W1906692317 creator A5001231520 @default.
• W1906692317 creator A5007273398 @default.
• W1906692317 creator A5011035523 @default.
• W1906692317 creator A5078220154 @default.
• W1906692317 creator A5082022450 @default.
• W1906692317 date "1996-12-01" @default.
• W1906692317 modified "2023-10-18" @default.
• W1906692317 title "Comparison of A60 and three glycolipid antigens in an ELISA test for tuberculosis" @default.
• W1906692317 cites W1883581227 @default.
• W1906692317 cites W1966946832 @default.
• W1906692317 cites W1990796481 @default.
• W1906692317 cites W1995809406 @default.
• W1906692317 cites W1996684511 @default.
• W1906692317 cites W2010744907 @default.
• W1906692317 cites W2010966423 @default.
• W1906692317 cites W2024873621 @default.
• W1906692317 cites W2036395450 @default.
• W1906692317 cites W2037520612 @default.
• W1906692317 cites W2038129611 @default.
• W1906692317 cites W2038202546 @default.
• W1906692317 cites W2040563840 @default.
• W1906692317 cites W2044640042 @default.
• W1906692317 cites W2053370638 @default.
• W1906692317 cites W2054605756 @default.
• W1906692317 cites W2066054086 @default.
• W1906692317 cites W2072136676 @default.
• W1906692317 cites W2077071189 @default.
• W1906692317 cites W2103224795 @default.
• W1906692317 cites W2110084454 @default.
• W1906692317 cites W2113714168 @default.
• W1906692317 cites W2128191049 @default.
• W1906692317 cites W2134990301 @default.
• W1906692317 cites W2145948175 @default.
• W1906692317 cites W2314003891 @default.
• W1906692317 cites W2339492498 @default.
• W1906692317 cites W4234405837 @default.
• W1906692317 cites W4312861399 @default.
• W1906692317 doi "https://doi.org/10.1016/s1198-743x(14)65145-4" @default.
• W1906692317 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/11866846" @default.
• W1906692317 hasPublicationYear "1996" @default.
• W1906692317 type Work @default.
• W1906692317 sameAs 1906692317 @default.
• W1906692317 citedByCount "28" @default.
• W1906692317 countsByYear W19066923172012 @default.
• W1906692317 countsByYear W19066923172014 @default.
• W1906692317 countsByYear W19066923172015 @default.
• W1906692317 countsByYear W19066923172016 @default.
• W1906692317 crossrefType "journal-article" @default.
• W1906692317 hasAuthorship W1906692317A5001231520 @default.
• W1906692317 hasAuthorship W1906692317A5007273398 @default.
• W1906692317 hasAuthorship W1906692317A5011035523 @default.
• W1906692317 hasAuthorship W1906692317A5078220154 @default.
• W1906692317 hasAuthorship W1906692317A5082022450 @default.
• W1906692317 hasBestOaLocation W19066923171 @default.
• W1906692317 hasConcept C142724271 @default.
• W1906692317 hasConcept C147483822 @default.
• W1906692317 hasConcept C151730666 @default.
• W1906692317 hasConcept C159047783 @default.
• W1906692317 hasConcept C203014093 @default.
• W1906692317 hasConcept C22222904 @default.
• W1906692317 hasConcept C2777267654 @default.
• W1906692317 hasConcept C2781069245 @default.
• W1906692317 hasConcept C71924100 @default.
• W1906692317 hasConcept C86803240 @default.
• W1906692317 hasConceptScore W1906692317C142724271 @default.
• W1906692317 hasConceptScore W1906692317C147483822 @default.
• W1906692317 hasConceptScore W1906692317C151730666 @default.
• W1906692317 hasConceptScore W1906692317C159047783 @default.
• W1906692317 hasConceptScore W1906692317C203014093 @default.
• W1906692317 hasConceptScore W1906692317C22222904 @default.
• W1906692317 hasConceptScore W1906692317C2777267654 @default.
• W1906692317 hasConceptScore W1906692317C2781069245 @default.
• W1906692317 hasConceptScore W1906692317C71924100 @default.
• W1906692317 hasConceptScore W1906692317C86803240 @default.
• W1906692317 hasIssue "3" @default.
• W1906692317 hasLocation W19066923171 @default.
• W1906692317 hasLocation W19066923172 @default.
• W1906692317 hasOpenAccess W1906692317 @default.
• W1906692317 hasPrimaryLocation W19066923171 @default.
• W1906692317 hasRelatedWork W1996768316 @default.
• W1906692317 hasRelatedWork W2013954651 @default.
• W1906692317 hasRelatedWork W2067648084 @default.
• W1906692317 hasRelatedWork W2087190059 @default.
• W1906692317 hasRelatedWork W2095837715 @default.
• W1906692317 hasRelatedWork W2281009136 @default.
• W1906692317 hasRelatedWork W2884171184 @default.
• W1906692317 hasRelatedWork W3048597464 @default.
• W1906692317 hasRelatedWork W4253812259 @default.
• W1906692317 hasRelatedWork W4281487762 @default.
• W1906692317 hasVolume "2" @default.
• W1906692317 isParatext "false" @default.
• W1906692317 isRetracted "false" @default.
• W1906692317 magId "1906692317" @default.
• W1906692317 workType "article" @default.