Why should this posting be reviewed?
See also Guidelines for Comments and Corrections.
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.close
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Reader Comments (1)
Post a new comment on this article
Rejection of TB Serodiagnosis by WHO. A Reply.
Posted by andamars on 07 Jan 2012 at 07:30 GMT
Rejection of TB Serodiagnosis by WHO. A Reply.
Roland Maes, Dr Sci.
Anda biologicals, Strasbourg, France
TB-clinicians use tests (sputum microscopy, nucleic acid amplification, culture) for the detection of the TB pathogen. The presence of the pathogen triggers a therapy. TB-Clinicians also use skin tests and interferon tests (an enzyme-linked serodiagnosis of interferon response to M. tuberculosis antigens) for the detection of latent infections. The detection of antibodies in association with antigen-detection tests, useful for diagnosis 1, 2, 3, 4, prognosis 3, 4, and to monitor illness 5, is currently practised only in India, mainly via the anda-TB tests.
The WHO pronounced a vigorous rebuttal of the serodiagnosis of tuberculous infections6. This rejection is based on an expert group meeting report (22 July 2010)7, itself based on a meta analysis8 of published evaluations of the accuracy and specificity of diagnostic tests, and on a cost-effectiveness study performed in India9. The quality of the published evaluations was rated by the GRADE system 10, 11.
The limitations of the studies analysed, the inconsistency of the results reported, the poor directness of the evidences, the imprecision and the bias uncovered by the analysts and evaluated by the panel, were judged of very low quality, implying that no further research would be likely to impact on confidence in the estimate of effect, and would likely change the estimate. The verdict was thus without appeal.
No scientist active in TB research in India was part of the panel who made the analysis. He could have argued that a comparison is systematically made between microscopy, taken as the gold standard, and the serological determination of presence of antibodies to mycobacterial antigens. The two methodologies are totally different and, in addition, the quality of evidence is never questioned for the gold standard itself, although its sensitivity varies from 20% to 80% in pulmonary cases, and of course is 0% in extra-pulmonary cases, which account for at least 30% of all TB-diseased subjects. He could have pointed out that the rejected studies were done according to the standards of those days. The majority of the investigators who made these studies had gone through adverse personal experiences. Adverse personal experience —what A. Stuebe12 calls Level IV evidence — is more convincing than even the tightest of confidence intervals. Informed adverse anecdote transforms scattered data into sound clinical judgment.
The rejected studies were Level II observational studies and Level III expert opinion. These levels are currently assimilated to Complementary Medicine in that they complement the recently advocated scientifically proven methods13. Chief among the hurdles facing the use of TB serology is a pervasive attitude that refers to it as “quackery.” However, more than half of the current recommendations of the Infectious Diseases Society of America are based on level III evidence only14. Anecdotal evidence is an inherent part of Western medicine.
The analysts restricted their meta-analysis to papers published since 1990. The initial studies verifying the merits of serology were published between 1998 and 1990 and the subsequent studies informing on enigmas and paradoxes15 arising from its use (inapparent infections in healthy populations16, 17, 18, 19 and in patients suffering from other diseases4, 20, 21, 22, anergy in subjects suffering from an immunosuppression5, 23, 24, 25, potentiation by BCG of mycobacterial infections as M. avium and M. tuberculosis 26, 27, 28, 29) were of sufficient strength for the serological test to be submitted to the Indian regulatory agency by a WHO Officer expert on TB in India. This officer and the agency were insightful in allowing a test that met the needs of the clinicians and patients in India.
The Panel experts’ report notes that 28 (42%) studies involved smear–negative patients. Smear-negativity is traced to abacillary cases, extra-pulmonary cases, primary tuberculosis in infants, and also to human errors (the sensitivity of smear microscopy ranges from 20% to 80% in overt pulmonary infections, explained by lack of electricity, lenses lost or dirty, stains corrupted, sputum contaminated or unsuitable, fatigue of the microscopist). Serological tests can go a long way in the detection of unapparent infections, prognostic and prevention. The immunopathology of TB is a complex phenomenon29, 30, 31, 32, 33 whose comprehension sheds light on the evolution of the disease.
Several members of the panel contributed to the development of the Elispot34, 35, 36, 37, 38, 39, 40, which is a serological test for the detection of a humoral interferon response to the mycobacterial antigen ESAT-6. The ESAT-6 antigens are associated with a risk of reactivation of latent TB and consequently are in direct competition with the anda-tb test applied as a prognostic tool 26, 41. An interferon test produced by the SDHO laboratory was found wanting42.
Another concern is raised by the SDHO laboratory located at Saint Sauveur des Monts, near Montreal. This SDHO laboratory is mentioned in table 2 -#17 of the panel report as a rapid test included in the meta-analysis, with two studies made (table 3), one used for the monitoring of HIV patients (page 23, § 3.1.6.). However, this rapid test was not included with the 19 rapid tests analysed by F. Portaels (table 4), the city of Saint Sauveur does not know the laboratory to exist and no information about the producers of the test were given to investigators42.
Conflict of interest is a major reproach risen in the report against the researchers who investigated the usefulness of serological tests, on the ground that kits were donated by manufacturers. This donation took also place for the analysis of rapid tests by WHO: “Companies interested in participating were asked to donate tests for evaluation”.
To claim repeatedly that the accuracy of investigated tests are even lower than reported because bad results would not have been published is unacceptable : the investigators report their results as they find them, be these findings favourable or not23, 24, 25, 42, 53.
I noted an inconsistency in the statement of Executive summary versus page 31:
--summary :(6) compared with ELISAs [60% (95% CI 6% to 65%], immuno-chromatographic assays had lower sensitivity [53%, 95% CI 42% to 64%];
--Page 31: (6) compared with ELISAs [60% (95% CI 6% to 65%], immuno-chromatographic assays have similar sensitivity [53%, 95% CI 42% to 64%];
The analysts considered all suitable publications appearing after 1990. However, the study of Charpin43, essential for the determination of the accuracy of the anda-TB serodiagnostic tests (IgM and IgG) and their usefulness and applications, was not included in the meta-analysis. The results of Gevaudan44 concerning pulmonary cases are not mentioned (“344 cases of primary pulmonary tuberculosis, 88% were positive for anti-A60 IgG and 75% for the corresponding IgM”) and only extra-pulmonary cases are considered (Fig 23 of the report) but no difference was drawn between IgG and IgM antibodies for the evaluation of the results, which confused the outcome. Similarly, the results obtained by Alifano45 are included in the meta-analysis, but not those reported in the same paper and a subsequent paper46 on extrapulmonary cases. Likewise, the study of Kaustova4, which shed light on unapparent mycobacterial infections among patients hospitalised for non-tuberculous diseases, was not considered. Unfortunately, the panellists limited their search to investigations performed from 1990 on, and some early investigations 2, 47, 48, 49, 50 were lost.
The meta analysis emphasizes that no tests were made on infants and children. See the following references51, 52, 53, 54, 55, 56.
The meta analysis emphasizes that only one rapid test (SDHO test) was made on HIV populations. The sensitivity of the SDHO test was 16% (95% CI 5% to 34%). Two other studies23,24 also report a poor sensitivity for the anda-Tb test and a third study25 reports a 50% sensitivity for the Anda-TB test. Production of humoral antibodies in HIV-infected subjects is of course rare because the immunodepression inflicted by HIV also affects TB: the negative results observed are not “false negatives” but “true negatives” traced to immunodepression.
Human errors account for the spread of results observed in microscopy (20% to 80%). Such errors cannot be excluded in studies on serology. Reference 5 of the report mentions a study25 that focused on the identification of latent infections. It reports: The sensitivity for detecting M. tuberculosis antibodies in human immunodeficiency virus-associated TB was 50% for both the InBios Active TbDetect ELISA and the ANDA Biologics TB ELISA and 0% for the IBL M. tuberculosis ELISA. Yet, all incubations for the Anda TB kit were performed at room temperature instead of the prescribed 37°C. This low and variable temperature considerably impacts the accuracy (much lower sensitivity and different speeds of development leading to spread of results in inter-essays). Focusing on latent infections, the publication says: “The Anda Biologicals TB ELISA detected 24 positive samples from this group, giving a positivity rate of 30.8%, about 10 times higher than the expected conversion rate (3 to 5%) from latent TB infection to active TB disease that has been cited in several studies on TB infection rates (3, 24).”.
The reference 3 is the guidelines of the American Thoracic Society of the year 200057 . These guidelines in turn base their own evaluation of latent infections on their reference 20, which reports a conversion rate observed in a study58 performed in 1968 in the UK on school children aged 14-19 years. This study reports that 4.7% of skin test positive subjects developed clinical TB within 15 yr of entry into the study. This evaluation of 4.7% is upgraded in reference 2425, which discloses59 that “the lifetime risk of reactivation tuberculosis is 20 percent or more among most persons with induration of 10 mm or more on a tuberculin skin test and either human immunodeficiency virus infection or evidence of old, healed tuberculosis”. This recent evaluation (20% or more of reactivation risk) reflects more closely the true level of the conversion rate. When correct conclusions are drawn from correctly performed investigations25, the Anda-tb test appears the best test to detect latent infections. The analysts included this study25 in their analysis and added chaos to an already chaotic situation.
B. cost/effectiveness :
The evaluation60 was performed on a hypothetical Indian population :“Since anda-TB is likely to outperform more poorly studied in-house serological tests and less accurate rapid test formats, and laboratory accuracy is likely to exceed that in the field, our analysis likely overestimates the accuracy of serology”. The 3 suppositions included in this single sentence render the conclusions of the cost analysis problematic. The most recent discussion on the correlation of cost with effectiveness61 states: “the approach provides a solid step forward in thinking about how to think about cost.”.
A myriad of publications exist on all aspects of TB. Heterodox views are not infrequently expressed in abstracts and non-peer-reviewed papers, which should be evaluated on their merits and not excluded on sentimental grounds: bad news are not necessary false. The chaos they have generated must be put in order by analyses and reviews whose first duty is to evaluate their worth, the second is to evaluate the validity of the conclusions drawn from correctly performed studies, and the third is to make correct recommendations based on these correct conclusions:
1. The regular stain-microscopy has zero TB-specificity.
2. BCG does not as a general rule elicit a delayed hypersensitivity reaction detected by skin tests.
3. BCG does not protect against TB, except in rare conditions (military TB in infants).
4. BCG favours super infection by other mycobacterial entities (M. avium, M. tuberculosis), and these pathogenic mycobacteria may induce a positive skin test. The skin tests observed after vaccination are due either to reactivation of latent M. tuberculosis or a superposed new infection.
5. MTB is immunodepressive by itself and favours infection by other immunodepressive pathogens as HIV, and reverse.
6. The humoral immune response to TB is just as important as the cellular one, and is subject to depression by the pathogen but the depression acts in a variable way on the different antibodies (IgG, IgM, IgA) elicited in individual patients. A heavy bacterial load may correspond to negative serology, and reverse.
7. One third of humanity is infected by a mycobacterial entity. Most mycobacterial infections are latent and are observed with variable frequency in non –tuberculous patients suffering from diverse diseases, and in healthy contacts. Not all diseases allow a mycobacterial-superposed infection.
8. An immunosuppressive drug may show anti-TB activity in vitro. In patients, the immunosuppression will favour the multiplication of the pathogen.
1. Maes R, Homasson J-P, Kubin M, Bayer M. Development of an enzyme immunoassay for the serodiagnostic of tuberculosis and mycobacterioses. (1989) Med. Microbiol. Immunol. 178: 323-335.
2. Baelden M-C, Vanderelst B, Dieng M, Prignot J, Cocito C. Serological analysis of human tuberculosis by an Elisa with mycobacterial Antigen 60. (1990), Scand. J. Infect. Dis. 22: 63-73.
3. Cocito C. Properties of mycobacterial antigen complex A60 and its applications to the diagnosis and prognosis of tuberculosis.(1991) Chest 100: 1687-1693.
4. Kaustova J. Serological IgG, IgM and IgA diagnosis and prognosis of mycobacterial diseases in routine practice. (1995) Eur. J. Med. Res. 1: 393-403.
5. Fadda G, et al. Serodiagnosis and followup patients with pulmonary tuberculosis by enzyme-linked immunosorbent assay.(1992) Eur. J. Epidemiol. 8(1): 81-87.
6. WHO: Commercial serodiagnostic tests for diagnosis of tuberculosis. Policy Statement (2011). ISBN 978 92 4 150 205 4. available via (www.who.int)
7. WHO TDR : Commercial serodiagnostic tests for diagnosis of tuberculosis expert group meeting report (22 July 2010)
8. Steingart KR, Flores LL, Dendukuri N, Schiller I, Laal S, et al. Commercial Serological Tests for the Diagnosis of Active Pulmonary and Extrapulmonary Tuberculosis: An Updated Systematic Review and Meta-Analysis. (2011) PLoS Med 8(8): e1001062. doi:10.1371/journal.pmed.1001062
9. Dowdy DW, Steingart KR, Pai M. Serological Testing Versus Other Strategies for Diagnosis of Active Tuberculosis in India: A Cost-Effectiveness Analysis. (2011) PLoS Med 8(8): e1001074. doi:10.1371/journal.pmed.1001074
10. G. Guyatt, A. Oxman, G. Vist, R Kunz, Y Falck-Ytter et al.: GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. (2008) BMJ Volume 336, 924-926.
11. A. Schünemann, A. Oxman, J. Brozek, P Glasziou, R. Jaeschke, et al. GRADE: grading quality of evidence and strength of recommendations for diagnostic tests and strategies. (2008) BMJ Volume 336, 1106-1110.
12. Alison M. Stuebe. Level IV Evidence — Adverse Anecdote and Clinical Practice. (2011) n engl j med 365; 1(7) pages 8-9.
13. J Wapner. Contemplating Complementary Medicine. Posted: (2011) November 2, PloS Blogs: November 04, 2011 | Diverse Perspectives on Science and Medicine.
14. D. Heun Lee, Ole Vielemeyer. Analysis of Overall Level of Evidence Behind Infectious Diseases Society of America Practice Guidelines. (2011) Arch Intern Med. 171(1):18-22.
15. Maes H, Causse J.E, Maes RF. Mycobacterial infections: Are the Observed Enigmas and Paradoxes Explained by Immunosuppression and Immunodeficiency? (1996) Med. Hypotheses 46:163-171.
16. Maes R. Incidence of inapparent active mycobacterial infections in France detected by an IgG serological test based on antigen 60. (1989) Med. Microbiol. Immunol. 178: 315-321
17. Wirrmann Ch. Public health application of a serological test for tuberculosis: study of the incidence of inapparent infections among the employees of an alsatian supermarket. (1990) Eur. J. Epidemiol. 6(3): 304-308.
18. Gutierrez J, Alados JC, Garcia F, Quiros E. IgG antibody levels to mycobacterial antigen 60 in a healthy population of the province of Granada. (1990) Rev. Clini. Esp. 187: 381-382.
19. Aziz N, Bukhari MH, Muneer M, Tayyab M, Chaudhry NA. Evaluation of anti-tuberculous antibodies in healthy contact and non-contact persons. (2006) Biomedica 22(jan-jun/Bio-2):31-36.
20. Zatla F, Petithory JC. L'Elisa avec l'antigène A60 dans le diagnostique de la tuberculose - Réactions croisées dans la leishmaniose viscérale. (1989) Techn. Biol. 5: 220-225
21. Couetdic G, Michel-Briand Y, Barale T. Anticorps anti-BCG (Antigène A60) au cours d'infections non mycobactériennes. (1990) Méd. Malad. Infect. 20(4): 174-176.
22. Ferroni A, Sermet-Gaudelus I, Le Bourgeois M, Pierre-Audigier C, Offredo C, Rottman M, Guillemot D, Bernède C, Vincent V, Berche P, Gaillard JL. Measurement of Immunoglobulin G against mycobacterial antigen A60 in patients with cystic fibrosis and lung infection due to Mycobacterium abscessus. (2005) Clin. Infect. Dis. 40:58-66.
23. van der Werf TS, et al. Sero-diagnosis of tuberculosis with A60 antigen enzyme-linked immunosorbent assay : failure in HIV-infected individuals in Ghana. (1992) Med. Microbiol. Immunol. 181: 71-76.
24. Pouthier F, et al. Anti-A60 immunoglobulin G in the serodiagnosis of tuberculosis in HIV-seropositive and seronegative patients. (1994) AIDS 8: 1277-1280.
25. Anderson BL, Welch RJ, Litwin CM. Assessment of three commercially available serologic assays for detection of antibodies to Mycobacterium tuberculosis and identification of active tuberculosis. (2008) Clin Vaccine Immunol. 15(11):1644-9. Epub 2008 Sep 30.
26. Ferreira P, Soares R, Arala-Chaves M. Susceptibility to infection with mycobacterium avium is paradoxically correlated with increased synthesis of specific anti-bacterial antibodies. (1991) Internat. Immunol. 3(5): 445-452.
27. Maes R. Clinical usefulness of serological measurements obtained by antigen A60 in
mycobacterial infections : development of a new concept. (1991) Klin. Wochenschr. 69: 696-709.
28. Maes RF. Tuberculosis II: the failure of the BCG vaccine. (1999) Med. Hypotheses 53(1): 32-39.
29. Maes RF. The tuberculosis enigma : need for a new paradigm importance of a knowledge of the immune status of the patients. (1999) Biomedicine 19(1): 1-14.
30. Harboe M, Closs O, Bjorvatn B, Bjune G. Antibodies against BCG antigen 60 in mycobacterial infection.( 1977) Brit. Med. J. 2: 430-433
31. Maes H, Causse JE, Maes R. Tuberculosis I: a conceptual frame for the immunopathology of the disease. (1999) Med. Hypotheses 52(6): 583-593.
32. Amicosante M, et al. Antibody repertoire against the A60 antigen complex during the course of pulmonary tuberculosis. (1993) Eur. Respir. J. 6: 816-822.
33. Carlucci S, et al. Mycobacterial antigen complex A60-specific T-Cell repertoire during the course of pulmonary tuberculosis. (1993) Infect. Immun. 61(2): 439-447.
34. Hill PC, Brookes RH, Adetifa IM, Fox A, Jackson-Sillah D, Lugos MD, Donkor SA,Marshall RJ, Howie SR, Corrah T, Jeffries DJ, Adegbola RA, McAdam KP. Comparison of enzyme-linked immunospot assay and tuberculin skin test in healthy children exposed to Mycobacterium tuberculosis. (2006) Pediatrics. 117(5): 1542-8.
35. Silva VM, Kanaujia G, Gennaro ML, Menzies D. Factors associated with humoral response to ESAT-6, 38 kDa and 14 kDa in patients with a spectrum of tuberculosis. (2003) Int J Tuberc Lung Dis.7(5): 478-84.
36. Hill PC, Brookes RH, Fox A, Jackson-Sillah D, Jeffries DJ, Lugos MD, Donkor SA,
Adetifa IM, de Jong BC, Aiken AM, Adegbola RA, McAdam KP. Longitudinal assessment of an ELISPOT test for Mycobacterium tuberculosis infection. (2007) PLoS Med. Jun;4(6):e192.
37. Hill PC, Jackson-Sillah D, Fox A, Franken KL, Lugos MD, Jeffries DJ, Donkor SA,
Hammond AS, Adegbola RA, Ottenhoff TH, Klein MR, Brookes RH. ESAT-6/CFP-10 fusion protein and peptides for optimal diagnosis of mycobacterium tuberculosis infection by ex vivo enzyme-linked immunospot assay in the Gambia. (2005) J Clin Microbiol. 43(5):2070-4.
38. Ifedayo MO Adetifa, Moses D Lugos, Abdulrahman Hammond, David Jeffries, Simon Donkor, Richard A Adegbola and Philip C Hill. Comparison of two interferon gamma release assays in the diagnosis of Mycobacterium tuberculosis infection and disease in The Gambia. (2007) BMC Infectious Diseases, 7:122
39. Aiken AM, Hill PC, Fox A, McAdam KP, Jackson-Sillah D, Lugos MD, Donkor SA,
Adegbola RA, Brookes RH. Reversion of the ELISPOT test after treatment in Gambian tuberculosis cases. (2006) BMC Infect Dis. 30;6:66.
40. Adetifa IM, Lugos MD, Hammond A, Jeffries D, Donkor S, Adegbola RA, Hill PC. Comparison of two interferon gamma release assays in the diagnosis of M. tuberculosis infection and disease in The Gambia. (2007) BMC Infect Dis. 25;7:122.
41. Wu HP, Hua CC, Yu CC, Wu SY. Comparison of plasma interferon-gamma and antigen 60 immunoglobulin G in diagnosing pulmonary Mycobacterium tuberculosis infection. (2005) Chang Gung Med J.
42. Kassa-Kelembho E, Zandanga G, Serrvice YB, Ignaleamoko A and A Talarmin: Poor performance of a novel serological test for the detection of pulmonary Tuberculosis in Bangui, Central Africa Republic. 6eme Confais, Dakar, 2006 Fais A-159. March 20-23, 2006.
43. Charpin D, et al. Value of ELISA using antigen A60 in the diagnosis of active pulmonary tuberculosis. (1990) Am. Rev. Respir. Dis. 142: 380-384.
44. Gevaudan M-J, Bollet C, Charpin D, Mallet MN, De Micco Ph. Serological response of tuberculosis patients to Antigen 60 of BCG. (1992) Eur. J. Epidemiol. 8(5): 666-676.
45. Alifano M, Del Pezzo M, Lambertini C, Faraone S, Covelli I. Elisa method for evaluation of anti-A60 IgG in patients with pulmonary and extrapulmonary tuberculosis. (1994) Microbiologica 17: 37-44.
46. Alifano M, et al. Detection of IgG and IgA against the mycobacterial antigen A60 in patients with extrapulmonary tuberculosis (1998) Thorax 53: 377-380.
47. Casal M, Linares MJ. Diagnostico serologico de la tuberculosis. Estudio de la primera prueba comercializada existente (AndaElisa-Tuberculosis).( 1988) Rev. Esp. Microbiol. Clin. 3: 491-492.
48. Gulletta E, Del Pezzo M, Sanduzzi A, Bariffi F, Covelli I. Serodiagnosis survey of tuberculosis by a new Elisa method. (1988) Eur. J. Epidemiol. 4(3): 331-334.
49. Raheman SF, Wagner S, Mauch H, Vasudeva ND, Ingole DL. Evaluation of a dual antigen Elisa test for the serodiagnosis of tuberculosis.(1988) Bull. W. H. O. 66(2): 203-209.
50. Mattar S, et al. Deteccion de anticuerpos IgG con el método "Elisa" an pacientes tuberculosos utilizando el antigeno 60. (1989) Rev. Esp. Microbiol. Clin. 2: 97-102.
51. Delacourt C, et al.Value of ELISA using antigen A60 for the diagnosis of tuberculosis in children. (1993) Chest 104: 393-398.
52. Rota S, Beyazova U, Karsligil T, Cevheroglu C. Humoral immune response against antigen 60 in BCG vaccinated infants. (1994) Eur. J. Epidemiol. 10: 713-718.
53. Turneer E, et al. Determination of Humoral Immunoglobulins M and G Directed against Mycobacterial Antigen 60 Failed to Diagnose Primary Tuberculosis and Mycobacterial Adenitis in Children. (1994) Am. J. Respir. Crit. Care. Med. 150: 1508-1512.
54. Singh P, Baveja CP, Talukdar B, Kumar S, Mathur MD. Diagnostic Utility of ELISA Test Using Antigen A60 in Suspected Cases of Tuberculous Meningitis in Paediatric Age Group. (1999) Indian. J. Pathol. Microbiol. 42(1): 11-14.
55. Swaminathan S, Umadevi P, Shantha S, Radhakrishnan A, Datta M. Sero diagnosis of tuberculosis in children using Two Elisa kits. (1999) Indian. J. Pediatr. 66: 837-842.
56. Khalilzadeh S, Yazdanpanah S. Sero diagnosis of tuberculosis in children using A60 antigens. (2001) Int. J. Tubercul. Lung. Dis. 5(11):200.
57. American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. (2000) Am. J. Respir. Crit. Care Med. 161:221–247.
58. Sutherland, I. The ten-year incidence of clinical tuberculosis following "conversion" in 2550 individuals aged 14 to 19 years. (1968) TSRU Progress Report (KNCV, The Hague, Netherlands).
59. Horsburgh, C. R., Jr. Priorities for the treatment of latent tuberculosis infection in the United States. (2004) N. Engl. J. Med. 350:2060–2067.
60. Dowdy DW, Steingart KR, Pai M. Serological Testing Versus Other Strategies for Diagnosis of Active Tuberculosis in India: A Cost-EffectivenessAnalysis. (2011) PLoS Med 8(8): e1001074. doi:10.1371/journal.pmed.1001074
61. Jessica Wapner: Taking a Stab at Cost-Effectiveness PloS blog Posted: April 6, 2011.