Doi:10.1016/s0140-6736(03)12950-9

Comparison of T-cell-based assay with tuberculin skin test for diagnosis of Mycobacterium tuberculosis infection in a school
tuberculosis outbreak
Katie Ewer, Jonathan Deeks, Lydia Alvarez, Gerry Bryant, Sue Waller, Peter Andersen, Philip Monk, Ajit Lalvani IntroductionIdentification and treatment of people who have latent Background The diagnosis of latent tuberculosis infection tuberculosis infection by targeted tuberculin skin testing relies on the tuberculin skin test (TST), which has many and preventive therapy is a cornerstone of tuberculosis drawbacks. However, to find out whether new tests are better control in developed countries.1 The main drawback of than TST is difficult because of the lack of a gold standard test the tuberculin skin test (TST) is poor specificity, since for latent infection. We developed and assessed a sensitive previous Mycobacterium bovis BCG vaccination and enzyme-linked immunospot (ELISPOT) assay to detect T cells environmental mycobacterial exposure can lead to false- specific for Mycobacterium tuberculosis antigens that are absent from Mycobacterium bovis BCG and most environmental tuberculosis in developed countries is carried by foreign- mycobacteria. We postulated that if the ELISPOT is a more born immigrants from high-prevalence countries, among accurate test of latent infection than TST, it should correlate whom BCG vaccination and environmental mycobacterial better with degree of exposure to M tuberculosis.
exposure are common.5,6 The TST also has severaloperational drawbacks, including the need for a return Methods A large tuberculosis outbreak in a UK school visit and operator-dependent variability in placement and resulted from one infectious index case. We tested 535 reading of the test. A more accurate rapid test for latent students for M tuberculosis infection with TST and ELISPOT.
infection is a major priority for improved tuberculosis We compared the correlation of these tests with degree of exposure to the index case and BCG vaccination.
The identification of genes in the M tuberculosis genome that are absent from M bovis BCG8 and most Findings Although agreement between the tests was high environmental mycobacteria9 offers an opportunity to (89% concordance, ␬=0·72, p<0·0001), ELISPOT correlated develop more specific tests for M tuberculosis infection.10 significantly more closely with M tuberculosis exposure than Early secretory antigen target-6 (ESAT-6) and culture did TST on the basis of measures of proximity (p=0·03) and filtrate protein 10 (CFP10) are two such gene products duration of exposure (p=0·007) to the index case. TST was that are strong targets of the cellular immune response in significantly more likely to be positive in BCG-vaccinated than tuberculosis patients and contacts.11,12 The presence of in non-vaccinated students (p=0·002), whereas ELISPOT ESAT-6-specific T cells, detected by the rapid ex-vivo results were not associated with BCG vaccination (p=0·44).
enzyme-linked immunospot (ELISPOT) assay forinterferon-gamma,13 is a highly sensitive and specific Interpretation ELISPOT offers a more accurate approach marker of M tuberculosis infection in patients who have than TST for identification of individuals who have latent culture-confirmed tuberculosis; its sensitivity is tuberculosis infection and could improve tuberculosis control substantially higher than that for the TST.14,15 In a UK by more precise targeting of preventive treatment.
pilot study of 50 contacts at risk of latent tuberculosisinfection, we noted a correlation between ESAT-6 ELISPOT results and the extent of exposure totuberculosis cases,16 whereas unexposed people wereuniformly ELISPOT-negative.17,18 In February, 2001, a secondary school student who had had a chronic cough for 9 months was diagnosed withsputum-smear-positive cavitatory pulmonary tuberculosis.
The health authority screened 1128 of 1208 students atthe school with TST and diagnosed 69 secondary cases ofactive tuberculosis and 254 cases of latent infection. Thisoutbreak presented a unique opportunity to compare theeffectiveness of the ELISPOT assay with the TST.
In the absence of a gold standard reference test, direct Nuffield Department of Clinical Medicine, University of Oxford, John Radcliffe Hospital, Oxford, UK (K Ewer BSc, L Alvarez PhD, assessment of the sensitivity and specificity of a new test A Lalvani DM); Centre for Statistics in Medicine, Institute of Health for latent tuberculosis infection is impossible.4 However, Sciences, Oxford, UK (J Deeks MSc); Leicestershire Health since airborne transmission of M tuberculosis is promoted Authority, Leicester, UK (G Bryant MFPHM, S Waller SRN, by increasing duration and proximity of contact with an P Monk FFPHM); and Statens Serum Institut, Copenhagen, Denmark infectious case,19–21 a key determinant of infection is the amount of time spent sharing room air with the source Correspondence to: Dr A Lalvani, Nuffield Department of Clinical case.22,23 We formed the hypothesis that if the ELISPOT Medicine, University of Oxford, Level 7, John Radcliffe Hospital, assay is a more sensitive and specific test than the TST, it should correlate more closely than the TST with degree of exposure to M tuberculosis and should be independent THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.
of BCG vaccination status. Two measures of exposure were prespecified at the time of study design: proximity to the index case, based on school class and year, and hours of direct classroom contact. Three features of this outbreak made it particularly suitable for this investigation: there was one infectious index case with several hundred contacts; the outbreak occurred in an enclosed environment; and school timetables permitted precise quantification of the amount of time each child Risk factors for tuberculosis exposure outside school spent sharing room air with the source case.
The study was approved by the Leicestershire research ethics committee. We invited 963 students, aged 11–15 years, from the same school as the index case to participate. We obtained written informed consent from 594 (62%) children and their parents. In May and June,2001, the school nurses interviewed 550 (57% of the total invited) of these children about place of birth and history Heaf grade (equivalent induration after 10 TU Mantoux) of tuberculosis exposure outside school. At the same time they drew 10 mL blood samples that were stored in sequentially numbered heparinised containers. Leicestershire Health Authority screened 1128 children with the Heaf test, in accordance with UK guidelines for tuberculosis contacts (table 1),24 535 of whom were in our sample of 550. Screening was done over 2 weeks, from March 26 to April 11, 2001, 2 months after the index case *India (44), Pakistan (nine), Bangladesh (three), Africa (Malawi, Kenya, and Tuberculin skin testing was done by standard multiple- Tanzania, 14), Portugal (two), and Greece, Malaysia, Sri Lanka, and Turkey (oneeach). puncture Heaf test with a six-needle disposable-head Heaf Table 1: Characteristics of students tested by TST and ELISPOT gun (Bignall Surgical Instruments, Littlehampton, UK)25and concentrated purified protein derivative 100 000tuberculin units per mL (Evans Medical, Liverpool, UK), had no access to personal identifiers or TST results. in accordance with national guidelines.24 Heaf tests were How ELISPOT assays are done has been previously administered and read by the medical and nursing staff of described;13,14 for this study we used a simplified, faster the outbreak management team. Cutaneous induration protocol incorporating ELISPOT plates precoated with was scored 1 week later, in accordance with standard monoclonal antibody to interferon-gamma (Mabtech AB, guidelines, from grade 0 to 4.25 Generally, although the Stockholm, Sweden), and a detector monoclonal antibody read-out of the automated Heaf test is quantified less to interferon-gamma preconjugated to alkaline-phosphatase precisely than the Mantoux test—ie, grades 0–4 instead of (Mabtech). Plates were seeded with 2·5ϫ105 peripheral mm of induration, a continuous variable—the two tests blood mononuclear cells per well: duplicate wells contained generally correlate well with each other.24–26 no antigen (negative control), phytohaemagglutinin Students who reported symptoms underwent chest (positive control; ICN Biomedicals, OH, USA), radiography and clinical assessment for possible active recombinant dimeric ESAT-6 (dimESAT-6), or one of 12 tuberculosis, irrespective of skin test results.
different peptide pools derived from ESAT-6 and CFP10.
Asymptomatic students with Heaf grades 0 or 1 or Heaf After overnight incubation at 37°C, 5% carbon dioxide , grade 2 and a BCG scar or documented history of BCG plates were developed with preconjugated detector vaccination were deemed uninfected24,25 and no action was antibody and chromogenic substrate, BCIP/NBTPLUS (Moss taken; students with Heaf grades 3 or 4 (irrespective of BCG vaccination history) or grade 2 with no evidence of Assays were scored by automated ELISPOT counter previous BCG vaccination were deemed infected.24,25 All (AID-GmbH, Strassberg, Germany). We scored test wells underwent chest radiography and those with normal as positive if they contained a mean of at least five more radiographs were deemed to have latent tuberculosis spot-forming cells than the mean of the negative control infection and received 3 months’ chemoprophylaxis with wells and this number was at least twice the mean of the rifampicin and isoniazid. Students with abnormal negative control wells. This cut-off14 was predefined radiographic findings or with symptoms were further before the results were revealed. Assays were deemed assessed in hospital for active tuberculosis; those with positive if there was a positive response to one or more positive cultures for M tuberculosis from clinical samples or pools of the ESAT-6-derived or CFP10-derived peptides, positive radiological or clinical findings suggestive of tuberculosis were classified as having active tuberculosis As previously described,14 we used peptides spanning disease. These students were treated with standard short- the length of ESAT-6 and CFP10 (ResGen, Huntsville, course chemotherapy for 6 months, including AL, USA). Each peptide was 15 aminoacids long and pyrazinamide and ethambutol for the first 8 weeks.
overlapped its adjacent peptide by 10 residues; purity was We did ELISPOT assays in Oxford on blood samples more than 70%. Peptides were arranged into 12 pools from 545 of the 550 students, 2–4 h after venepuncture.
comprising two arrays of six pools each, where each array Samples were processed and scored by two scientists who contained all 35 peptides from both molecules in THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.
vacated by the index case as indirectlyexposed. Direct and indirect exposure and increase as one or both of thesemeasures increase. For latent estimate test sensitivity and testspecificity directly, but were able to each test to the likelihood of infection.
We estimated the increase in odds increase in exposure by logisticregression. We used matched-pairslogistic regression to assess the Figure 1: TST and ELISPOT results for students stratified by decreasing proximity to index case based on school year and class T+=TST positive. T–=TST negative. E+=ELISPOT positive. ELISPOT–=ELISPOT negative. A: students in same class as index case. B: students in classes in same year who regularly shared lessons with index case. C: students in the four remaining classes in same year who shared only weekly school events but no lessons with index case. D: students in different years who shared no contrasting combinations, so that each peptide was tested BCG vaccination, place of birth, and household tuberculosis contact. All reported p values are two sided.
We cloned, expressed, and purified DimESAT-6 from We investigated trends with use of the ␹2 statistic.
culture supernatant of recombinant Lactococcus lactis; Comparisons between proportions were derived with Fisher’s exact test. We did all analyses with STATA(version 7.0).
Ascertainment of exposureWe classified school students into four groups of decreasing degrees of exposure to the index case, based on The sponsors of the study had no role in the study proximity and shared activities in school: the same class as design, data collection, data analysis, data interpretation, the index case; students in classes in the same year (year 9) writing of the report, or in the decision to submit the who regularly shared classes with the index case; students in classes in the same year who shared only weekly eventswith the index case; and students in different years (7, 8, and 10) who shared no school events with the index case ELISPOT and TST results were available for 535 (figure 1). For students in the same school year, we used students—44·3% of the school. Our sample was the school timetable to quantify direct exposure to the representative in terms of the proportion of non-white index case, taking into account the attendance record of children (97% in our sample vs 93% in the whole the index case during the likely infectious period, which, on the basis of duration of cough and associated diagnosed with active tuberculosis (5 vs 6%); and symptoms, was 9 months. Since the index case mixed with participants deemed to have latent tuberculosis infection different students for each academic subject, substantial on the basis of TST result (24 vs 23%, table 1).
numbers of students were exposed. We classified students The odds of a test result being positive for each from other school years (years 7, 8, and 10) who had increase across the four stratified exposure groups lessons in classrooms immediately after they had been increased by a factor of 2·78 (95% CI 2·22–3·48, p for TST
vs ELISPOT
Exposure to M tuberculosis in school
Stratified exposure groups (whole school, n=535)
Direct exposure (weeks) in year 9 (n=148) Indirect exposure (weeks) in years 7, 8, and 10 (n=387) Risk factors for exposure to M tuberculosis outside school
History of household tuberculosis contact (n=36)
*Positive result defined as Heaf grade >2 or heaf grade 2 without BCG.
Table 2: Odds ratios (95% CI) of the relations of ELISPOT and TST with intensity of M tuberculosis exposure in school and with risk
factors for exposure outside school
THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.
p for vaccinated vs
*␹2 test for trend across all five Heaf grades.
Table 3: Effect of previous BCG vaccination on ELISPOT andTST results p<0·0001) for the ELISPOT assay and 2·33 (1·88–2·88, p<0·0001) for the TST. The ELISPOT assay correlated Figure 2: Stratification of TST-positive students with significantly better with increasing exposure across the presumed latent tuberculosis infection by ELISPOT result four groups than did the TST (p=0·03; figure 1, table 2). 128 TST-positive students with no clinical or radiographical signs of Direct exposure of the 148 children in year 9 ranged active tuberculosis disease. Each circle represents one student:white=students with direct classroom exposure to index case; from 0 to 17 school weeks: 57 students had some direct classroom exposure, with a median of 2·2 school weeks(IQR 1·4–13·4). The odds of a positive ELISPOT result (p=0·002), with substantially more Heaf grade 3 (81 of increased by a factor of 2·51 (1·58–3·99, p<0·0001) with 467 vs 2 of 68, p=0·001), and grade 2 results in BCG- each week of direct exposure, which was significantly higher (p=0·007) than that for the TST (odds ratio 1·30 Of the 128 participants presumed to have latent [95% CI 1·10–1·54], p=0·002; table 2). tuberculosis infection on the basis of a positive TST Of the 387 children in years 7, 8, and 10, 196 pupils with no evidence of active tuberculosis, 97 (76%) had indirect exposure, up to a maximum of 1·16 weeks.
tested positive with ELISPOT. This ELISPOT-positive Although ELISPOT and TST were more likely to be subgroup had significantly higher Heaf grades and positive with increasing exposure, neither showed a significantly more exposure to M tuberculosis than did the ELISPOT-negative students. Heaf grades were ELISPOT assay and TST were positively correlated significantly higher among ELISPOT-positive students with a history of household tuberculosis contact (n=36, than among ELISPOT-negative students (p<0·0001, table 2). 76 children were born in countries with a high figure 2). In the ELISPOT-positive group there were prevalence of tuberculosis and climates associated with significantly more students with direct exposure to the increased exposure to environmental mycobacteria (table index case than in the ELISPOT-negative group (35 of 1). The mean duration of residence in these countries 97 vs one of 31, p<0·0001; figure 2).
was 7·8 years. TST results, but not ELISPOT results, Agreement between TST and ELISPOT was high were significantly associated with birth in one of these (␬=0·72 [95% CI 0·64–0·80], p<0·0001), with concordant results in 475 (89%) students (table 4). For For 362 students, the date of BCG vaccination was students in whom test results were discordant, it is documented in the Leicestershire Health Authority impossible to know for certain which test was correct records, of whom 323 were vaccinated at birth. An because there is no reference test. However, table 4 additional 105 students had BCG scars but the date of shows that an isolated positive ELISPOT result (ie, one vaccination was not available because they were born associated with a negative TST) was a strong predictor of outside Leicester; 101 were born in countries where BCG M tuberculosis exposure, whereas an isolated positive TST vaccination is given at birth. Therefore 424 (91%) of 467 result was not. This finding suggests that isolated positive BCG-vaccinated students were vaccinated at birth. The ELISPOT results are more likely to be true positives than ELISPOT assay showed no significant relation with BCG are isolated positive TST results. For students with vaccination status (p=0·44, table 3). By contrast, BCG- positive TST and ELISPOT results, the relative risk of vaccinated children were significantly more likely to have direct exposure to the index case, compared with that for higher Heaf grades than unvaccinated children students with negative TST and ELISPOT, was 17·6 Students’ exposure to M tuberculosis
pTST–/ELISPOT+ vs TST– ELISPOT–* TST+/ELISPOT– vs TST– ELISPOT–* NT=no statistical test undertaken due to skewed distributional shape because most students had no direct exposure. *p values are for comparison of epidemiologicalcharacteristics of students with an isolated positive test result vs characteristics of students negative for both tests.
Table 4: Analysis of concordant and discordant test results THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.
The high specificity of ELISPOT might explain the strong relation between positive ELISPOT results andTST induration size in individuals who have positive TSTresults. The size of the TST response is positivelyassociated with higher tuberculosis case rates duringfollow-up; thus, the ELISPOT assay may have identifiedthe subgroup of TST-positive individuals who actuallyhave latent tuberculosis infection. These individuals aredistinct from those whose weakly positive TST responsesrepresent false-positive results due to antigenic crossreactivity of purified protein derivative. Moreover, theELISPOT-positive group had substantially more exposureto M tuberculosis than did the ELISPOT-negative group.
This improved specificity of the ELISPOT could help toavoid unnecessary chemoprophylaxis in uninfectedindividuals; this ability to screen out false-positive TSTresults will become increasingly important as theprevalence of latent infection falls in low-prevalence Figure 3: Restriction fragment length polymorphism patterns of M tuberculosis isolates from students at school
derivative may explain why a new whole-blood interferon- Top row is isolate from index case; other nine rows are isolates from gamma ELISA based on purified protein derivative secondary cases from who M tuberculosis was cultured.
There is compelling evidence that the outbreak we (95% CI 8·1–38·0, p<0·001); for those with negative studied was due to one index case, who was the first TST and positive ELISPOT results it was 11·7 symptomatic case of pulmonary tuberculosis in the (4·2–33·2, p<0·001); and for those with postive TST and school.30 The molecular epidemiology also suggests that negative ELISPOT results, it was 2·97 (0·6–13·7, this was a point-source outbreak. Only two other children Molecular strain typing by variable-number tandem symptomatic for less than 2 weeks before admission to repeat, mycobacterial interspersed repetitive unit, and hospital and both were in year 11, which did not spoligotyping showed that all nine secondary isolates of participate in the study. Moreover, their M tuberculosis M tuberculosis from students at the school were identical isolates were identical to that of the index case by all four to that of the index case. IS6110-based restriction fragment length polymorphism (figure 3) showed that The high rates of tuberculosis infection and disease at seven of the nine secondary isolates were identical to that the school are unlikely to merely reflect the epidemiology of the index case, whereas two were very similiar, of tuberculosis in the local community. First, this outbreak differing by one and three bands each.
accounted for a third of all tuberculosis cases in Leicesterin 2001. Second, all 1226 household contacts of the 69 tuberculosis cases and 254 cases of latent tuberculosis In the absence of a gold standard test for latent infection were screened by the health authority and no tuberculosis infection, the sensitivity and specificity of the cases of infectious pulmonary tuberculosis were identified.
ELISPOT assay or the TST cannot be directly Third, four other Leicester schools were screened by TST, quantified.4 However, given that the likelihood of latent and the rates of positive skin tests were 1–4%. Fourth, tuberculosis infection is determined by exposure to when year 8 students at this school underwent Heaf testing M tuberculosis,19–23 we were able to rank the tests according to their diagnostic accuracy. Agreement between TST and The minimum exposure to an infectious person that is ELISPOT results was high, but discordance in 11% of required for M tuberculosis transmission is unknown, but students shows that the tests are not equivalent. Our must be low, since many well-documented cases of results indicate that ELISPOT probably has higher infection result from brief exposure31 and many students sensitivity and specificity than TST. First, the significantly who did not share lessons with the source case must have closer correlation of ELISPOT than TST with degree of acquired infection in this way. The amount of exposure exposure to M tuberculsosis suggests a higher sensitivity for required before transmission of M tuberculosis becomes detection of latent infection. Second, TST, but not inevitable is also unknown. Since all students with 5 or ELISPOT, was confounded by BCG vaccination, despite more school-weeks of exposure had positive results on the 11–15 years having elapsed since vaccination, which ELISPOT assay, however, our findings suggest that 130 h suggests a higher specificity for the ELISPOT assay. sharing room air with a person with sputum smear- TST and ELISPOT were more likely to be positive in positive cavitatory tuberculosis is certain to result in students who had a history of household tuberculosis contact, a marker of M tuberculosis exposure outside Longitudinal assessment of the positive predictive school, than in students without such a history. By value of this assay for subsequent development of active contrast, for the students born in high-prevalence tuberculosis will be necessary. In one report workers countries, mainly Africa and Asia (a risk factor for suggest that T-cell responses to ESAT-6 in healthy environmental myobacterial exposure3,4 and M tuberculosis contacts are associated with subsequent active disease.32 exposure) only the TST was significantly more likely to be Students in our study who had positive ELISPOT but positive. Given that the ELISPOT assay correlates negative TST results, who have not had chemoprophylaxis, strongly with all other measures of M tuberculosis are receiving close clinical and radiographic follow-up.
exposure, its independence from foreign birth suggests We used the Heaf test, because it is used for tuberculin that, unlike TST, it is not confounded by environmental testing in contact investigations in the UK, and is stipulated in national guidelines.24 Since the Mantoux THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.
method is more widely used internationally, ELISPOT Taylor Z, O’Brien RJ. Tuberculosis elimination: are we willing to pay should be compared in the future with this method; we the price? Am J Respir Crit Care Med 2001; 163: 1–2.
have recently started such studies in several countries. Mahairas GG, Sabo PJ, Hickey MJ, Singh DC, Stover CK. Molecular ELISPOT gives quantitative results the morning after analysis of genetic differences between Mycobacterium bovis BCG and
virulent M bovis. J Bacteriol 1996; 178: 1274–82.
taking a blood sample and is more convenient, objective, Harboe M, Oettinger T, Wiker HG, Rosenkrands I, Andersen P.
and rapid than the TST. Although TST is cheap, related Evidence for occurrence of the ESAT-6 protein in Mycobacterium indirect costs are associated with return visits and the tuberculosis and virulent Mycobacterium bovis and for its absence in trained staff required to administer and read the test.
Mycobacterium bovis BCG. Infect Immun 1996; 64: 16–22.
10 Andersen P, Munk ME, Pollock JM, Doherty TM. Specific immune- Introduction of ELISPOT might initially increase the cost based diagnosis of tuberculosis. Lancet 2000; 356: 1099–104.
of tuberculosis control, but the savings that would follow 11 Lein AD, von Reyn CF, Ravn P, Horsburgh CR Jr, Alexander LN, from improved diagnosis of latent tuberculosis infection Andersen P. Cellular immune responses to ESAT-6 discriminate could make it very cost effective in the long term. Better between patients with pulmonary disease due to Mycobacterium avium detection of latent infection would lessen the number of complex and those with pulmonary disease due to Mycobacterium
tuberculosis. Clin Diagn Lab Immunol 1999; 6: 606–09.
cases of active tuberculosis and, therefore, the attendant 12 Arend SM, Andersen P, van Meijgaarden KE, et al. Detection of cost of diagnosis, hospital admission and contact tracing.
active tuberculosis infection by T cell responses to early-secreted Fewer false-positive results in uninfected contacts antigenic target 6-kDa protein and culture filtrate protein 10. would avoid the costs associated with unnecessary J Infect Dis 2000; 181: 1850–54.
chemoprophylaxis and its associated toxic effects.
13 Lalvani A, Brookes R, Hambleton S, Britton WJ, Hill AV, McMichael AJ. Rapid effector function in CD8+ memory T cells.
J Exp Med 1997; 186: 859–65.
14 Lalvani A, Pathan AA, McShane H, et al. Rapid detection of Ajit Lalvani, Katie Ewer, Jonathan Deeks, Gerry Bryant, and Philip Monk Mycobacterium tuberculosis infection by enumeration of antigen- designed the study. Ajit Lalvani coordinated the study. Katie Ewer and specific T cells. Am J Respir Crit Care Med 2001; 163: 824–28.
Lydia Alvarez did the ELISPOT assays. Sue Waller interviewed and enrolled 15 Barnes PF. Diagnosing latent tuberculosis infection: the 100-year all the participants. Demographic information was obtained and recorded by upgrade. Am J Respir Crit Care Med 2001; 163: 807–8.
Sue Waller, Gerry Bryant, and Philip Monk. Jonathan Deeks computed the 16 Lalvani A, Pathan AA, Durkan H, et al. Enhanced contact duration of exposure and did the statistical analysis. Peter Andersen tracing and spatial tracking of Mycobacterium tuberculosis infection synthesised the recombinant ESAT-6 and provided technical advice and by enumeration of antigen-specific T cells. Lancet 2001; 357:
support. Ajit Lalvani wrote the paper with help from Katie Ewer and Jonathan Deeks, and all researchers reviewed the final report.
17 Lalvani A, Nagvenkar P, Udwadia Z, et al. Enumeration of T cells specific for RD1-encoded antigens suggests a high prevalence of latent Mycobacterium tuberculosis infection in healthy urban Indians. AL is the named inventor on several patents related to T-cell-based J Infect Dis 2001; 183: 469–77.
diagnosis filed by the University of Oxford since 1996. Regulatory 18 Chapman AL, Munkanta M, Wilkinson KA, et al. Rapid detection of approval and commercialisation of the ELISPOT assay will be undertaken active and latent tuberculosis infection in HIV-positive individuals by by a spin-out company of the University of Oxford (Oxford Immunotec), enumeration of Mycobacterium tuberculosis-specific T cells. AIDS in which AL has an equity stake. PA is the named inventor of several 2002; 16: 2285–93.
patents filed by Statens Serum Institute relating to the discovery of M 19 Grzybowski S, Barnett GD, Styblo K. Contacts of cases of active tuberculosis-specific antigens.
pulmonary tuberculosis. Bull Int Union Tuberc 1975; 50: 90–106.
20 Stead WW. Undetected tuberculosis in prison. Source of infection for community at large. JAMA 1978; 240: 2544–47.
We thank the students for their participation; Gary Coleby, Alan Anderson, 21 Kenyon TA, Valway SE, Ihle WW, Onorato IM, Castro KG.
Isobel Pearce, Dean Barnett and the staff of Crown Hills Community Transmission of multidrug-resistant Mycobacterium tuberculosis during College for their generous cooperation; Lorna Briars, Elaine Weare, a long airplane flight. N Engl J Med 1996; 334: 933–38.
Alison Woodbridge, and Fiona Booth for interviewing and drawing blood 22 Houk VN, Baker JH, Sorensen K, Kent DC. The epidemiology of samples from the students; Debbie Modha, Judith West, Wren Hoskyns, tuberculosis infection in a closed environment. Arch Environ Health Louise Coole, Helen Thuraisingham, Ginder Narle, Niru Rana, 1968; 16: 26–35.
Lindsey Abbott, Vicki Lowe, and Barbara Smithson for epidemiological 23 Houk VH, Kent DC, Baker JH, Sorensen K, Hanzel GD. The Byrd information from the outbreak investigation; Francis Drobniewski and study: in-depth analysis of a micro-outbreak of tuberculosis in a Malcolm Yates of the Mycobacterial Reference Library, London, for RFLP closed environment. Arch Environ Health 1968; 16: 4–6.
typing; John Watson of the Communicable Disease Surveillance Centre for 24 Joint Tuberculosis Committee of the British Thoracic Society.
facilitating the project and valuable advice; Doug Altman, Constantine Control and prevention of tuberculosis in the United Kingdom: code Gatsonis, Patrick Bossuyt, and Les Irwig for guidance in planning the of practice 2000. Thorax 2000; 55: 887–901.
statistical analysis; Peter Barnes, Luca Richeldi, Peter Wrighton-Smith, 25 Department of Health. Immunisation against infectious disease.
Shabbar Jaffar, Paul Newton, Liz Corbett, and Anthony Butterworth for London: H M Stationery Office, 1996.
helpful discussions; Ramilla Mistry and Dina Shah for translations; and 26 Carruthers KJ. Comparison of the Heaf (multiple puncture) Ansar Pathan, Katalin Wilkinson and Tilly Pillay for laboratory assistance.
and Mantoux tests using several tuberculins. Tubercle 1969; 50:
This study was funded by the Wellcome Trust. AL is a Wellcome senior research fellow in clinical science.
27 Deeks JJ. Systematic reviews of evaluations of diagnostic and screening tests. In: Egger M, Davey-Smith G, Altman D, eds.
Systematic reviews in health care: meta-analysis in context. London: Targeted tuberculin testing and treatment of latent tuberculosis infection: joint statement of the American Thoracic Society (ATS) and 28 Brock I, Munk ME, Kok-Jensen A, Andersen P. Performance of the Centers for Disease Control and Prevention (CDC). whole blood IFN-gamma test for tuberculosis diagnosis based on Am J Respir Crit Care Med 2000; 161: S221–47.
PPD or the specific antigens ESAT-6 and CFP-10. Int J Tuberc Lung Huebner RE, Schein MF, Bass JB, Jr. The tuberculin skin test. Dis 2001; 5: 462–67.
Clin Infect Dis 1993; 17: 968–75.
29 Mazurek GH, LoBue PA, Daley CL, et al. Comparison of a whole- Black GF, Weir RE, Floyd S, et al. BCG-induced increase in blood interferon gamma assay with tuberculin skin testing for interferon-gamma response to mycobacterial antigens and efficacy of detecting latent Mycobacterium tuberculosis infection. JAMA 2001; BCG vaccination in Malawi and the UK: two randomised controlled 286: 1740–47.
studies. Lancet 2002; 359: 1393–401.
30 Press releases. http://www.leics-ha.org.uk (accessed March 18, 2003).
Jasmer RM, Nahid P, Hopewell PC. Clinical practice: latent 31 Small PM, Hopewell PC, Singh SP, et al. The epidemiology of tuberculosis infection. N Engl J Med 2002; 347: 1860–66.
tuberculosis in San Francisco: a population-based study using Zuber PL, McKenna MT, Binkin NJ, Onorato IM, Castro KG. conventional and molecular methods. N Engl J Med 1994; 330:
Long-term risk of tuberculosis among foreign-born persons in the United States. JAMA 1997; 278: 304–07.
32 Doherty TM, Demissie A, Olobo J, et al. Immune responses to the Rose AM, Watson JM, Graham C, et al. Tuberculosis at the end of Mycobacterium tuberculosis-specific antigen ESAT-6 signal subclinical the 20th century in England and Wales: results of a national survey in infection among contacts of tuberculosis patients. J Clin Microbiol 1998. Thorax 2001; 56: 173–79.
2002; 40: 704–60.
THE LANCET • Vol 361 • April 5, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group.

Source: http://listas.lampada.uerj.br/lampada/ementas/aulas/epidemio/Tb_diagnostico.pdf