Detection of immunoglobulin G antibodies to Toxoplasma gondii: Evaluation of two commercial immunoassay systems

Transcription

1 Journal of Microbiology, Immunology and Infection (211) 44, 21e26 available at journal homepage: ORIGINAL ARTICLE Detection of immunoglobulin G antibodies to Toxoplasma gondii: Evaluation of two commercial immunoassay systems Nilgun Tekkesin a, *, Kenan Keskin b, Cumhur Kılınc a, Nuray Orgen a, Kaya Molo a a Department of Biochemistry, Clinical Laboratory, Memorial Hospital, Istanbul, Turkey b Department of Microbiology and Disease, Clinical Laboratory, Memorial Hospital, Istanbul, Turkey Received 12 August 29; received in revised form 2 October 29; accepted 11 November 29 KEYWORDS Anti-T gondii IgG; Immunofluorescent assay; Toxo IgG Background: Toxoplasmosis is a disease, which can cause severe congenital infection and is normally diagnosed by the detection of Toxoplasma gondii (T gondii)-specific antibodies in the serum of infected patients. Several different tests allow to distinguish recent from past infections and to quantify anti-t gondii-specific IgG, and the results can be used as markers for a chronic or recently seroconverted toxoplasma. Methods: In the present study, the recent Cobas 6 Toxo IgG assay (Roche Diagnostics, Indianapolis, IN, USA) for the serological diagnosis of toxoplasmosis was compared with the Axsym Toxo IgG assay (Abbott Laboratories, Diagnostics Division, Abbott Park, IL, USA) employing a panel of negative, low- or high-reactive serum samples that were selected after routine screening in a laboratory of clinical analyses. Results: The overall agreement between two methods was 99% (r Z.99, p <.1). Of 91 analyzed samples, only one presented discrepant result, being positive in the Cobas 6 Toxo IgG assay and negative in the Axsym Toxo IgG assay. By using an immunofluorescent assay as a confirmation test, this positive result was assayed to be negative. Conclusions: Both assays performed in each analyzer were proven to be fast and fully automated procedures for reproducible measurement of IgG antibodies to T gondii. The new method, used for the determination of anti-t gondii IgG antibodies, should be evaluated with a further analysis with increased number of serum samples to get a broad performance of this newer test. Copyright ª 211, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. All rights reserved. * Corresponding author. Department of Biochemistry, Clinical Laboratory, Memorial Hospital, Piyalepasa Bulvari 34385, Istanbul, Turkey. address: niltek@hotmail.com (N. Tekkesin) /$36 Copyright ª 211, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. All rights reserved. doi:1.116/j.jmii

2 22 N. Tekkesin et al. Introduction Toxoplasmosis, a ubiquitous protozoal disease, is caused by an intracellular parasite, Toxoplasma gondii (T gondii). This is mainly acquired by ingestion of food or water that is contaminated by mature oocysts shed by cats or by undercooked meat containing tissue cysts. 1 Primarily, acute toxoplasmosis, which is mostly mild or asymptomatic, is generally benign in healthy people. However, it can be serious in the context of immunodeficiency, especially in the case of AIDS and bone marrow or heart transplant patients or in children infected in utero. However, reactivation of a latent Toxoplasma as a result of immunosuppression is frequently associated with meningoencephalitis. 2,3 Primary maternal toxoplasma occurs during pregnancy may lead to severe damage of the fetus as the parasite can transmit across the placenta. Most infants with congenital infection do not present any clinical symptoms at birth but may develop severe sequelae later in life, such as mental and psychomotor retardation, chorioretinitis, and hear loss. 4 The fetal rate increases with gestational age. However, the risk of severe clinical manifestations is higher in case of early maternal infection. 4e6 Early drug therapy in acute toxoplasmosis during pregnancy can prevent congenital damage or ameliorate the severity of clinical manifestations. 4e6 A wide range of assays for the diagnosis of toxoplasma is available on the commercial market, and many are used in laboratories throughout Turkey. The assays vary in their format and in their ability to detect different types of antibody classes (some IgG specifically, others all antibody classes). This feature may lead to variation in the performance of the assays and makes the choice of assay for individual diagnostic laboratories complex. In addition, laboratories should use evidence-based medicine in their decision-making process. In recent years, clinical laboratories have seen a drastic increase in demand for automated detection of antibodies to T gondii by assays based on immunoassay technology. 7,8 The determination of IgG antibodies with a significant rise titers to T gondii is used to assess the serological status to toxoplasmosis and indicate a chronic or recently seroconverted toxoplasma. In the present study, the performance characteristics of the new Toxo IgG antibody assay on the Cobas 6 analyzer (Roche Diagnostics, Indianapolis, IN, USA) was compared with those of Axsym Toxo IgG (Abbott Laboratories, Diagnostics Division, Abbott Park, IL, USA) antibody assay on the Axsym automated analyzer systems with respect to reproducibility and correlation. We review the performance of these assays and offer the importance of the results in patient counseling for specialist and nonspecialist laboratories. Methods The study related to 91 sera. They were obtained from the routine activity of the Central Clinical Laboratory at Memorial Hospital for the period MarcheApril, 27 and consisted of sera from pregnant women, newborns, immunocompetent, and immunocompromised patients. They included negative samples as well as samples with low, medium, and high levels of IgG antibodies to T gondii detected by Toxo IgG Axsym. Cobas 6 Toxo IgG, marketed by Roche Diagnostics, is a relatively new product but increasingly being used in Europe. The sera were stored at 2 C until testing. For each serum tested by Axsym Toxo IgG before, the biological interpretation (presence or absence of IgG) considered results were obtained with Cobas 6 Toxo IgG. Cobas 6 Toxo IgG antibody immunoassay Ten microliter sample, a biotinylated recombinant T gondii-specific antigen and T gondii-specific recombinant antigen labeled with a ruthenium complex form a sandwich complex. After addition of streptavidin-coated microparticles, the complex bind to the solid phase via interaction of biotin and streptavidin. The reaction mixture is aspirated into the measuring cell where the microparticles are magnetically captured onto the surface of the electrode. Unbound substances are then removed with ProCell. Application of a voltage to the electrode then induces chemiluminescent emission, which is measured by a photomultiplier. Results are expressed as international units per milliliter in the IgG assay. Interpretation of IgG results was based on the manufacturer s criteria as follows: 3. IU/mL, positive for IgG antibodies; 1. to <3. IU/ ml, equivocal result; and <1. IU/mL, negative for IgG antibodies. Abbott Axsym Toxo IgG antibody immunoassay The Abbott Axsym uses an automated system, which is based on microparticle enzyme immunoassay technology to detect antibodies to T gondii. After manual pipetting of sera into individual reaction cells, all subsequent steps are automated. The instrument delivers the sample and diluent buffers to the predilution well of the reaction cell. T gondii-coated microparticles and diluted sample are added to the incubation well, where the T gondii antibody binds to the T gondii-coated microparticles, forming an antigenantibody complex. A diluted aliquot of this antigen-antibody complex is transferred to the glass fiber matrix, which irreversibly binds the microparticles. After washing the matrix to remove unbound materials, class-specific antihuman IgG-alkaline phosphatase conjugate is dispensed onto the matrix, where it binds to the antigen-antibody complex. The matrix is again washed to remove unbound materials; this is followed by addition of the substrate, 4- methylumbelliferyl phosphate. The resulting fluorescent product is measured by the microparticle enzyme immunoassay optical assembly. The intensity of the fluorescence is proportional to the quantity of toxoplasma antibodies. Results are expressed as international units per milliliter. Interpretation of IgG results was based on the manufacturer s criteria as follows: >3. IU/mL, positive for IgG antibodies; 2.e3. IU/mL, equivocal result; and <2. IU/ ml, negative for IgG antibodies. System precisions as indicated by the coefficient of variation (CV%) were evaluated by intra- and interassay testing of negative and positive controls in each assay. Intra-assay precision was determined by running three consecutive runs of each negative and positive control for

3 Toxoplasma gondii, IgG detection 23 IgG antibodies. Interassay precision was determined by testing one negative and one positive control, each in duplicate, for 1 days. Results for sera with discrepant results for IgG antibodies were then verified by an immunofluorescent assay (Panbio, Inc., Columbia, MD, USA). System precisions as indicated by the CV% were evaluated by intra- and interassay testing of negative and positive controls in each assay. Statistics The CV% was used to determine intra- and interassay precision and reproducibility. CV% was used to determine the statistical agreement between the different tests using the McNemar test. Overall agreement was defined as the percentage of specimens that were positive or negative in the Axsym assay and gave the same positivity and negativity in the Cobas 6 assay. 9 Results The within-run and between-run reproducibility rates, calculated as the CV%, were between nonapplicable and 2.8% for serum samples tested by Axsym Toxo IgG and between 1.6% and 3.% for those tested with the Cobas 6 Toxo IgG assay. The results are presented in Table 1. The within-run reproducibility rates of negative and positive controls of each were below.1 for Axsym Toxo IgG and calculated CV% value was shown as nonapplicable. This indicated that Axsym Toxo IgG assay has a higher degree of assay and/or technical precision rather than Cobas 6 Toxo IgG assay. The assays proved to be highly reproducible both within runs and between runs. After successfully completing reproducibility testing, evaluation of the clinical specimens began. In the study comparing the Cobas 6 and Axsym Toxo IgG assays, 17 sera of 91 routine laboratory sera were considered as positive; and for anti-t gondii IgG, 73 as negative and 1 as discordant. Figures 1 and 2 show a graphic representation of the reactivity variations for negative and positive samples. Table 1 shows the results of 17 serum samples that were positive in both automated assays. The reactivity indices showed that the reactivity was significantly higher in the Cobas 6 when compared with the Axsym. The average reactivity of the Cobas 6 assay was 32.4 UI/mL, whereas the Axsym assay showed an average reactivity of 54.5 UI/mL. The reproducibility studies were also the same as the mean value of Cobas 6. Toxo IgG was higher than Axsym Toxo IgG (48.6 IU/mL vs IU/mL for positive controls and.767 IU/mL vs..1 IU/mL for negative controls). Only one serum sample has a discrepant result. It was Cobas 6 Toxo IgG assay positive (titer, 41 IU/mL) but Axsym Toxo IgG assay negative (titer, IU/mL). This sample was retested in each analyzer and the same results were obtained. The same serum was negative by indirect fluorescent antibody (IFA) test. The overall agreement between two methods was 99% (r Z.99, p <.1). These results indicate a close quantitative correlation between Axsym Toxo IgG assay and the Cobas 6 Toxo IgG assay. Table 2 illustrates the comparative performances of the Cobas 6 and Axsym assays. Discussion The detection of anti-t gondii IgG is considered to be a marker for a chronic or recently seroconverted toxoplasma and is very important as a reference for clinical and epidemiological patient management and counseling. Today, in the case of toxoplasmosis, new diagnostic parameters, such as IgG avidity, are assessed to identify the most likely time of and to allow the physician to better evaluate the clinical management of each patient. 1e12 In many health care services, the detection of anti-t gondiispecific IgG implicates that the patient is considered protected against the disease in case of a new contact with the parasite. Consequently, the performance of assays that detect specific IgG is of vital importance, and several investigations on this diagnostic approach are underway. A number of techniques have been applied to detect IgG antibodies in patients with toxoplasma, including the dye test, 13 direct agglutination assay, 14,15 IFA, 16 and ELISA. 17 Because of the practicability, rapidity, and reliability of immunoassays, automated detection of antibodies to T gondii on the basis of immunoassay technology has gained widespread use over the past few years in clinical laboratories. 8 Among these tools, the automated and standardized assay for anti-t gondii-specific IgG detection newly developed by Roche Diagnostics occupies an important position. Evaluation of clinical specimens for IgG-specific antibodies to T gondii was performed by using the Cobas 6 and Axsym assays. The Cobas 6 is a new product but is increasingly being used in Turkey. The Axsym Toxo IgG assay Table 1 Summary of precision data for Axsym and Cobas 6 Toxo IgG assays Axsym Toxo IgG Cobas 6 Toxo IgG Serum sample Mean (IU/mL) Within-run CV (n Z 6) a Between-run CV (n Z 14) b Mean (IU/mL) Within-run CV (n Z 6) Negative control.1 c NA Positive control 19.3 c NA a Three replicates per run. b Two samples per run for 1 days. c p <.1. CV Z coefficient of variation; NA Z nonapplicable. Between-run CV (n Z 14)

4 24 N. Tekkesin et al COBAS 6 (UI/mL) Sera positivity for Toxo G AXSYM (UI/mL) Figure 1. Positive reactivity of serum samples in each system. has been used throughout Turkey for many years. The Cobas 6 assay had not previously been evaluated with the Axsym assay. We review the performance of these assays and offer the best advice for specialist and nonspecialist laboratories in Turkey. Detection and measurement of IgGspecific antibodies is rarely problematic, and good sensitivity and specificity have been achieved by a variety of methods. 18e22 A clinical laboratory must not rely on the manufacturer to ensure a continuing supply of reagents but must institute its own verification procedures for each kit as well as performing an initial evaluation. 23 Kit inserts have demonstrated a sensitivity of 99.7% and specificity of 99.1% for Abbott Axsym and 99.5% and 98.8%, respectively for Cobas 6. As we could not have any chance to verify the other samples qualitatively by IFA, it would not be true to calculate the sensitivities and specialties of both assays. Further analysis should be organized for this purpose. Overall, the agreement between two assays was found very high (r Z.99, p <.1). Precision rates are generally presented as the assay precision rate but actually reflect the combination of the technologist s precision and that of the assay. With both assays, the addition of a measured amount of undiluted patient serum to the test cassette is the only manual procedure required. One would therefore expect improved precision with the automated assays as opposed to the manual assays. The Cobas 6 Toxo IgG is proved to be highly reproducible both within runs and between runs, with a higher detection titer than Axsym Toxo IgG in positive and negative reactive samples. Low titers of anti-t gondii IgG antibodies often maintain ambiguity on the protective immunity notion. In this regard, the high reactivity performance of Cobas 6 Toxo IgG assay might have an advantage. In this study, the material analyzed consisted of both negative and positive IgG serum samples and included a large range of positive IgG titers. The different IgG titers obtained from the same serum samples with different tests Sera negativity for Toxo G 1,25 1,25 1 1,75,75,5,5,25,25 COBAS 6 (UI/mL) AXSYM (UI/mL) Figure 2. Negative reactivity of serum samples in each system.

5 Toxoplasma gondii, IgG detection 25 Table 2 Comparison of performances of Axsym and Cobas 6 Toxo IgG assays a No. of serum samples (n Z 91) No. of serum samples with indicated results % Agreement Axsym Toxo IgG negative Axsym Toxo IgG positive Cobas 6 negative a Cobas 6 positive 1 17 a r Z.99 between two assays (p <.1). emphasize the variability in the quantitative results according to the serological methods. It is well established that identical titers cannot be obtained with commercial kits using different antigenic extracts from T gondii. 24 In studies that compare the performances of various methods to detect IgG antibodies to T gondii, it has been shown that no method is perfect. 25e27 Such studies have shown that certain sera yield discordant results. In our study, the results indicate close quantitative agreement between Axsym Toxo IgG assay and the Cobas 6 Toxo IgG assay. Compared with Axsym Toxo IgG assay, Cobas 6 showed only one discrepant result, confirmed by IgG-specific IFA test. The same sample was analyzed for the second time by another Cobas 6 Toxo IgG kit perhaps resulting variability among kit lots. But, again it was positive, probably because of a cross-reactivity, an undefined interference or may be explained by differences in the sensitivity and in the kinetics of these two methods. This suggests that the result obtained for this discrepant serum by the Cobas 6 Toxo IgG assay is false-positive. For this study, an immunofluorescent assay for T gondii IgG antibody was designated as the reference assay for the determination of discrepancy reason because of the depth of experience and data accumulated with this assay. 28 Sera collected during various stages of and convalescence often could give different results because methods have different sensitivities, measure different immunoglobulin classes of antibodies, or measure antibodies to different viral components. 29 One of the reasons for these test differences may be the different antigens coated on the solid phase of each assay and the resulting ability of each assay to detect these immunoglobulin classes and viral components. Specimens that yielded discordant test results following evaluation by Cobas 6 may be explained by such differences in solid-phase coating. The remaining discrepant Toxoplasma IgG specimen had insufficient volume for further testing. Otherwise, it was evaluated further by the Sabin-Feldman dye test 13 that would be the second step confirmatory analysis. The authors determined that the dye test should be retained as a reference test and as an additional confirmatory test for the validation of commercial kits. 3 As the results were obtained using retrospective samples, we have no chance to take a subsequent sample afterward. In a practical way, when positive results are obtained with the two methods, a subsequent serum sample taken 3 weeks later is tested, so as to verify the presence of a positive and stable titer of anti-t gondii IgG antibodies. In fact, the host defense mechanisms against Toxoplasma depend mainly on cellmediated immunity. 31 On the other hand, in case of discordant results, it is advisable to consider the patient as nonimmune, thereby necessitating a serological follow-up during the pregnancy. As we could not have any chance to verify the other samples qualitatively by IFA, it would not be true to calculate the sensitivities and specialties of both assays. Besides, Cimon et al. 32 found a lower sensitivity of IFA compared with Toxo IgG Axsym and may be related to the lack of antigen affinity purification and the sometimesdelicate microscope reading at the threshold dilution. Also, serological methods are often not sufficient in the followup of immunocompromised patients. 33 For Toxo IgG Axsym and IFA, Roux-Buisson et al. 34 found false-negative and false-positive results in immunocompromised patients. In practice, IFA should be used in a laboratory in combination with another method with different antibody kinetics in necessity. The knowledge of the specificity and sensitivity of test performance is fundamental for a physician so that he or she can make correct decisions and patient counseling, especially during pregnancy. A further analysis with a much larger sample of specimens should be done to evaluate a broad performance of this newer test. However, the results of this study should be taken into consideration when interpreting a positive result obtained with Cobas 6 Toxo IgG kit. References 1. Montoya JG, Liesenfeld O. Toxoplasmosis. Lancet 24;363: 1965e Luft BJ, Remington JS. Toxoplasmic encephalitis in AIDS. Clin Infect Dis 1992;15:211e Khalifa K el-s, Roth A, Roth B, Arasteh KN, Janitschke K. Value of PCR for evaluation occurrence of parasitemia in immunocompromised patients with cerebral and extracerebral Toxoplasmosis. J Clin Microbiol 1994;32:2813e6. 4. Remington JS, McLeod R, Desmonts G. Toxoplasmosis. In: Remington JS, Klein JO, editors. Infectious diseases of the fetus and newborn infant. 5th ed. Philadelphia, PA: W.B. Saunders; 21. p. 25e Thulliez P. Maternal and foetal infection. In: Joynson DHM, Wreghitt TG, editors. Toxoplasmosis: a comprehensive clinical guide. Cambridge, UK: Cambridge University Press; 21. p. 193e Wong SY, Remington JS. Toxoplasmosis in pregnancy. Clin Infect Dis 1994;18:853e9. 7. Bessieres MH, YLe Fichoux, Luyasu V, Chretien MC. Qualitative detection of IgA antibody to Toxoplasma gondii in a multicenter evaluation using the IMx analyzer; 1995:1355. abstr. Proceedings of the VIIth European Congress of Clinical Microbiology and Infectious Diseases. 8. Schaefer LE, Dyke JW, Meglio FD, Murray PR, Crafts W, Nile AC. Evaluation of microparticle enzyme immunoassays for

6 26 N. Tekkesin et al. immunoglobulin G and M to rubella virus and Toxoplasma gondii on the Abbott IMx automated analyzer. J Clin Microbiol 1989;27:241e3. 9. Ilstrup D. Statistical methods in microbiology. Clin Microbiol Rev 199;3:219e Ashburn D, Joss AW, Pennington TH, Ho-Yen DO. Do IgA, IgE and IgG avidity tests have any value in the diagnosis of toxoplasma in pregnancy? J Clin Pathol 1998;51:312e Camargo ME. Toxoplasmose. In: Ferreira AW, Ávila SLM, editors. Diagnóstico Laboratorial das principais doenças infecciosas e auto-imunes. Rio de Janeiro, Brazil: Guanabara Koogan; 21. p. 278e Cozon GJN, Ferrandiz J, Nebhi H, Wallon M, Peyron F. Estimation of the avidity of immunoglobulin G for routine diagnosis of chronic Toxoplasma gondii in pregnant women. Eur J Clin Microbiol Infect Dis 1998;17:32e Desmonts G. Sur la technique de lyse des toxoplasmes. Réaction de Sabin et Feldman. [The technical analysis of Toxoplasma. Reaction of Sabin Feldman]. Sem Hop Paris 1955;31: 193e5 [in French]. 14. Desmonts G. Dépistage de la toxoplasmose par agglutination des parasites. intére^t d un antige ne tre s sensible pour la recherche des IgG spécifiques. [Screening for toxoplasmosis by agglutination parasites. Interest of a antigenes very sensitive for specific IgG]. Ann Biol Clin 1983;41:5e138 [in French]. 15. Fulton JD, Turk JL. Direct agglutination test for Toxoplasma gondii. Lancet 1959;2:168e Garin JP, Ambroise-Thomas P. Le diagnostic sérologique de la toxoplasmose par la méthode des anticorps fluorescents. [The serological diagnosis of toxoplasmosis by the fluorescent antibody method]. Presse Med 1963;71:2485e8 [in French]. 17. Turunen H, Vuorio KA, Leinikki PO. Determination of IgG, IgM and IgA antibody responses in human toxoplasmosis by enzymelinked immunosorbent assay (ELISA). Scand J Infect Dis 1983; 15:37e Bacigalupo MA, Bazzini P, Farina L, Ius A. Evaluation of three immunoassays for detection of Toxoplasma-specific immunoglobulin G and M. Eur J Clin Chem Clin Biochem 1996;34: 53e Hayde M, Salzer HR, Gittler G, Aspock H, Pollak A. Microparticle enzyme immunoassay (MEIA) for Toxoplasma specific immunoglobulin G in comparison to the Sabin-Feldman dye test. A pilot study. Wien Klin Wochenschr 1995;17:133e6. 2. Liesenfeld O, Press C, Flanders R, Ramirez R, Remington JS. Study of Abbott Toxo IMx system for detection of immunoglobulin G and immunoglobulin M Toxoplasma antibodies: value of confirmatory testing for diagnosis of acute toxoplasmosis. J Clin Microbiol 1996;34:2526e Olsen MA, Root PP. Comparison of four different immunoassays for detection of Toxoplasma-specific immunoglobulin G. Diagn Microbiol Infect Dis 1994;19:19e Sandin RL, Knapp CC, Hall GS, Washington JA, Rutherford I. Comparison of the Vitek Immunodiagnostic Assay System with an indirect immunoassay (Toxostat Test Kit) for detection of immunoglobulin G antibodies to Toxoplasma gondii in clinical specimens. J Clin Microbiol. 1991;29:2763e National Committee for Clinical Laboratory Standards. Specifications for immunological testing for infectious diseases. Approved guideline. NCCLS document I/LA18-A. Wayne, PA: National Committee for Clinical Laboratory Standards; Derouin F, Garin YJF, Buffard C, Berthelot F, Petithory JC. Groupe de travail Toxoplasmose du Contro^ le National de Qualité en Parasitologie. Etude multicentrique de la sérologie toxoplasmique par différents réactifs ELISA commercialisés. [Working control of Toxoplasma National Quality Parasitology. Multicenter study of toxoplasma serology by ELISA reagents different markets?]. Bull World Health Organ 1994;72:7e249 [in French]. 25. Fuccillo D, Madden D, Tzan N, Sever J. Difficulties associated with serological diagnosis of Toxoplasma gondii. Diagn Clin Immunol 1987;5:8e Violand S, Mitchell T, Kleeman K. Comparison of an enzyme-linked immunoassay and a quantitative indirect fluorescent-antibody test with the conventional indirect fluorescent-antibody test for detecting antibodies to Toxoplasma gondii. J Clin Microbiol 1982; 16:341e Walton B, Benchoff B, Brooks W. Comparison of the indirect fluorescent antibody test and methylene blue dye test for detection of antibodies to Toxoplasma gondii. Am J Trop Med Hyg 1966;15:149e Wilson M, Ware DA, Walls KW. Evaluation of commercial serodiagnostic kits for toxoplasmosis. J Clin Microbiol 1987;25: 2262e Schaefer LE, Dyke JW, Meglio FD, Murray PR, Crafts W, Niles AC. Evaluation of microparticle enzyme immunoassays for immunoglobulins G and M to rubella virus and Toxoplasma gondii on the Abbott IMx automated analyzer. J Clin Microbiol 1989;27:241e3. 3. Reiter-Owona I, Petersen E, Joynson D, Aspöck H, Dardé ML, Disko R, et al. The past and present role of the Sabin-Feldman dye test in the serodiagnosis of toxoplasmosis. Bull World Health Organ 1999;77:929e Hunter CA, Suzuki Y, Subauste CS, Remington JS. Cells and cytokines in resistance to Toxoplasma gondii. Curr Top Microbiol Immunol 1996;219:113e Cimon B, Marty P, Morin O, Bessieres MH, Marx-Chemla C, Gay- Andrieu F, Thulliez P. Specificity of low anti-toxoplasma IgG titers with IMx and Axsym Toxo IgG assays. Diagn Microbiol Infect Dis 1998;32:65e Arnold SJ, Kinney MC, McCormick MS, Dummer S, Scott MA. Disseminated toxoplasmosis: Unusual presentations in the immunocompromised host. Arch Pathol Lab Med 1997;121: 869e Roux-Buisson N, Fricker-Hidalgo H, Foussadier A, Rolland D, Suchel-Jambon AS, Brenier-Pinchart MP, et al. Comparative analysis of the VIDAS Toxo IgG IV assay in the detection of antibodies to Toxoplasma gondii. Diagn Microbiol Infect Dis 25;53:79e81.