Interferon-c release assays for the diagnosis of latent Mycobacterium tuberculosis infection: a systematic review and meta-analysis R. Diel, D. Goletti, G. Ferrara, G. Bothamley, D. Cirillo, B. Kampmann, C. Lange, M. Losi, R. Markova, G.B. Migliori, A. Nienhaus, M. Ruhwald, D. Wagner, J.P. Zellweger, E. Huitric, A. Sandgren and D. Manissero ABSTRACT: We conducted a systematic review and meta-analysis to compare the accuracy of the QuantiFERON-TB1 Gold In-Tube (QFT-G-IT) and the T-SPOT1.TB assays with the tuberculin skin test (TST) for the diagnosis of latent Mycobacterium tuberculosis infection (LTBI). The Medline, Embase and Cochrane databases were explored for relevant articles in November 2009. Specificities, and negative (NPV) and positive (PPV) predictive values of interferon-c release assays (IGRAs) and the TST, and the exposure gradient influences on test results among bacille Calmette–Gue ´rin (BCG) vaccinees were evaluated. Specificity of IGRAs varied 98–100%. In immunocompetent adults, NPV for progression to tuberculosis within 2 yrs were 97.8% for T-SPOT1.TB and 99.8% for QFT-G-IT. When test performance of an immunodiagnostic test was not restricted to prior positivity of another test, progression rates to tuberculosis among IGRA-positive individuals followed for 19–24 months varied 8–15%, exceeding those reported for the TST (2–3%). In multivariate analyses, the odd ratios for TST positivity following BCG vaccination varied 3–25, whereas IGRA results remained uninfluenced and IGRA positivity was clearly associated with exposure to contagious tuberculosis cases. IGRAs may have a relative advantage over the TST in detecting LTBI and allow the exclusion of M. tuberculosis infection with higher reliability. KEYWORDS: ECDC, interferon-c release assay, latent Mycobacterium tuberculosis infection, meta-analysis, systematic review, TBNET I mprovement of diagnostic methods for latent Mycobacterium tuberculosis infection (LTBI) is an important step towards the goal of tuber- culosis elimination, as laid out by the WHO Stop TB strategy [1] and The European Centre for Disease Prevention and Control (ECDC) Frame- work Action Plan to Fight TB in the European Union [2]. As part of reaching this goal, indivi- duals infected with M. tuberculosis need to be identified and offered preventive therapy to stop the progression to active tuberculosis and prevent further M. tuberculosis transmission [3]. Thus, there is a need to develop more accurate methods for the detection of LTBI, and to provide evidence- based guidance on the use of such methods before they can be adopted by national tuberculosis screening programmes [4, 5]. In most areas of Europe, the identification of LTBI relies on the tuberculin skin test (TST). This diagnostic test has been assessed comprehen- sively in terms of its potential and limitations for use in preventive strategies for tuberculosis elimination [6]. However, the TST does not discriminate between potential infection with M. tuberculosis and prior vaccination with the bacille Calmette–Gue ´rin (BCG), or possible infec- tion with nontuberculous mycobacteria (NTM). Interferon (IFN)-c release assays (IGRAs) are in vitro immune tests that have been introduced in recent years as an alternative to the TST for the diagnosis of LTBI. IGRAs are based on the detection of a T-cell immune response towards M. tuberculosis complex specific antigens (early secretory antigenic target (ESAT)-6, culture fil- trate protein (CFP)-10 and/or TB7.7). To date, there are two commercially available platforms that measure IFN-c production following ex vivo antigen stimulation [7]: in the QuantiFERON1-TB AFFILIATIONS A full list of the authors’ affiliations can be found in the Acknowledgements section. CORRESPONDENCE C. Lange TBNET c/o Division of Clinical Infectious Diseases Medical Clinic Research Center Borstel Parkallee 35 23845 Borstel Germany E-mail: [email protected]Received: July 20 2010 Accepted after revision: Sept 13 2010 First published online: Sept 16 2010 European Respiratory Journal Print ISSN 0903-1936 Online ISSN 1399-3003 88 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL Eur Respir J 2011; 37: 88–99 DOI: 10.1183/09031936.00115110 CopyrightßERS 2011
12
Embed
Mycobacterium tuberculosis - European Respiratory Journal · expressed by the M. tuberculosis complex, with the exception of Mycobacterium kansasii, Mycobacterium szulgai, Mycobacterium
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Interferon-c release assays for the
diagnosis of latent Mycobacterium
tuberculosis infection: a systematic review
and meta-analysisR. Diel, D. Goletti, G. Ferrara, G. Bothamley, D. Cirillo, B. Kampmann, C. Lange,M. Losi, R. Markova, G.B. Migliori, A. Nienhaus, M. Ruhwald, D. Wagner,J.P. Zellweger, E. Huitric, A. Sandgren and D. Manissero
ABSTRACT: We conducted a systematic review and meta-analysis to compare the accuracy of
the QuantiFERON-TB1 Gold In-Tube (QFT-G-IT) and the T-SPOT1.TB assays with the tuberculin
skin test (TST) for the diagnosis of latent Mycobacterium tuberculosis infection (LTBI).
The Medline, Embase and Cochrane databases were explored for relevant articles in November
2009. Specificities, and negative (NPV) and positive (PPV) predictive values of interferon-c release
assays (IGRAs) and the TST, and the exposure gradient influences on test results among bacille
Calmette–Guerin (BCG) vaccinees were evaluated.
Specificity of IGRAs varied 98–100%. In immunocompetent adults, NPV for progression to
tuberculosis within 2 yrs were 97.8% for T-SPOT1.TB and 99.8% for QFT-G-IT. When test
performance of an immunodiagnostic test was not restricted to prior positivity of another test,
progression rates to tuberculosis among IGRA-positive individuals followed for 19–24 months varied
8–15%, exceeding those reported for the TST (2–3%). In multivariate analyses, the odd ratios for TST
positivity following BCG vaccination varied 3–25, whereas IGRA results remained uninfluenced and
IGRA positivity was clearly associated with exposure to contagious tuberculosis cases.
IGRAs may have a relative advantage over the TST in detecting LTBI and allow the exclusion of
M. tuberculosis infection with higher reliability.
Improvement of diagnostic methods for latentMycobacterium tuberculosis infection (LTBI) isan important step towards the goal of tuber-
culosis elimination, as laid out by the WHO StopTB strategy [1] and The European Centre forDisease Prevention and Control (ECDC) Frame-work Action Plan to Fight TB in the EuropeanUnion [2]. As part of reaching this goal, indivi-duals infected with M. tuberculosis need to beidentified and offered preventive therapy to stopthe progression to active tuberculosis and preventfurther M. tuberculosis transmission [3]. Thus,there is a need to develop more accurate methodsfor the detection of LTBI, and to provide evidence-based guidance on the use of such methods beforethey can be adopted by national tuberculosisscreening programmes [4, 5].
In most areas of Europe, the identification ofLTBI relies on the tuberculin skin test (TST). This
diagnostic test has been assessed comprehen-sively in terms of its potential and limitations foruse in preventive strategies for tuberculosiselimination [6]. However, the TST does notdiscriminate between potential infection withM. tuberculosis and prior vaccination with thebacille Calmette–Guerin (BCG), or possible infec-tion with nontuberculous mycobacteria (NTM).
Interferon (IFN)-c release assays (IGRAs) are invitro immune tests that have been introduced inrecent years as an alternative to the TST for thediagnosis of LTBI. IGRAs are based on thedetection of a T-cell immune response towardsM. tuberculosis complex specific antigens (earlysecretory antigenic target (ESAT)-6, culture fil-trate protein (CFP)-10 and/or TB7.7). To date,there are two commercially available platformsthat measure IFN-c production following ex vivoantigen stimulation [7]: in the QuantiFERON1-TB
88 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
Eur Respir J 2011; 37: 88–99
DOI: 10.1183/09031936.00115110
Copyright�ERS 2011
Gold In-Tube test (QFT-G-IT) and its predecessor theQuantiFERON1-TB Gold (QFT-G) test (Cellestis Ltd,Carnegie, Australia), the amount of IFN-c released into thesupernatant is quantified using an ELISA [8], whereas theproportion of blood cells releasing IFN-c is determined withthe T-SPOT1.TB assay (Oxford Immunotec Ltd, Abingdon,UK) using an enzyme-linked immunospot (ELISPOT) techni-que [9]. As the antigens used in IGRAs are almost exclusivelyexpressed by the M. tuberculosis complex, with the exception ofMycobacterium kansasii, Mycobacterium szulgai, Mycobacteriummarinum and Mycobacterium riyadhense, IGRAs are less likely tobe confounded by prior BCG vaccination and/or exposure toNTM [10].
For any new test to replace the TST, evidence of the test’shigher diagnostic accuracy is needed. Importantly, such a testshould show a higher specificity (ability to exclude M.tuberculosis infection), especially in subjects with confoundingfactors, such as BCG vaccination, and at least a similarsensitivity. However, the lack of a gold standard for diagnos-ing LTBI is a problem when trying to determine the diagnosticaccuracy of both the IGRAs and the TST [11]. An alternativeway to evaluate sensitivity, which cannot be assessed directlyin this context, is through comparison of the risk of developingactive tuberculosis in subjects with an increased risk ofprogression from LTBI to active disease.
The adoption and implementation of IGRAs into nationaltuberculosis screening programmes should be evidence-based.Several countries have, to date, adopted IGRAs for thediagnosis of LTBI within tuberculosis screening programmesand guidance documents on their use are available [4, 12, 13].Clinical guidance for the diagnosis of LTBI is especiallyimportant for the care of immunocompromised individuals,e.g. patients undergoing tumour necrosis factor (TNF)-antago-nists therapies [14], who carry an increased risk of progressionto active tuberculosis. Several studies, and consequent system-atic reviews and meta-analyses have been conducted evaluat-ing IGRA accuracy in diagnosing LTBI [15–18]. Furthermore,evidence on the actual predictive value of the IGRAs (i.e. therelationship between having a positive or negative IGRA resultand developing or not developing tuberculosis) is emerging[19–22]. To further support public health programmes, acontinuous assessment of the use of IGRA for the diagnosisof LTBI is, therefore, needed.
This article provides important up-to-date information on theuse of IGRAs to support the development of evidence-basedguidance for tuberculosis screening programmes. These datamay also be helpful for conducting more reliable cost-effectiveness and cost–benefit analyses with respect to theuse of IGRA or TST, solely or in combination.
Specifically, the diagnostic specificity of IGRAs relative to theTST, the IGRAs’ negative (NPV) and positive (PPV) predictivevalues for the progression to active tuberculosis, the associa-tion between exposure to patients with tuberculosis, and IGRAand TST test results in contacts were evaluated.
We present a systematic review and meta-analysis of theliterature according to evidence-based highest-standard cri-teria on the accuracy of commercially available IGRAs for thediagnosis of LTBI.
METHODSWe conducted this systematic review according to the guide-lines of the Preferred Reporting Items for Systematic Reviewsand Meta-Analyses statement [23] and the Quality Assessmentof Diagnostic Accuracy Studies (QUADAS) checklist [24].
Search strategy and selection criteriaWe searched Medline, Embase and the Cochrane CentralRegister of Controlled Trials (CENTRAL; Cochrane Library2009, issue 3) through November 15, 2009 for studies publishedin English. The following search terms were used:‘‘Tuberculosis’’ OR ‘‘TB infection’’ OR ‘‘TB disease’’ AND‘‘QuantiFERON’’ OR ‘‘Elispot’’ OR ‘‘T-SPOT’’ OR ‘‘interferon-gamma assay’’ OR ‘‘interferon-c release assay’’ OR ‘‘T cellassay’’ AND ‘‘ESAT-6’’ OR ‘‘CFP-10’’, were separatelycombined with the terms ‘‘NPV’’ OR ‘‘negative predictivevalue’’ OR ‘‘negative PV’’, ‘‘PPV’’ OR ‘‘positive predictivevalue’’ OR ‘‘predictive’’ and ‘‘TST’’ AND ‘‘BCG’’.
No restrictions were made with respect to basic study designor data collection (prospective or retrospective). Secondaryreferences cited by the studies and review articles retrievedfrom the databases were reviewed. Unpublished sources ofdata were not included in this systematic review.
From studies identified as potentially relevant, original articlesor letters to the editor were chosen that met the followingcriteria: original data were presented; the most recentcommercially available IGRAs (QFT-G-IT and T-SPOT1.TB)were used with European cut-offs for positive test results (inUS Food and Drug Administration-approved criteria, a personis T-SPOT1.TB negative if they have four or fewer spots abovethe nil control, positive if they have eight or more spots abovethe nil control, and borderline (or equivocal) if they have fiveto seven spots above the nil control); active disease inparticipants was confirmed by culture, nucleic acid amplifica-tion technique and/or histopathological examination; and, ifcomparing results of two or more tests, these had to beperformed in the same individuals.
The following studies were excluded: case reports, editorials,immunological (laboratory) and animal studies; studies notfollowing the manufacturers’ instructions, e.g. in vitro incuba-tion .24 h or freezing of cells; studies involving patientswhose diagnosis of active tuberculosis relied solely on clinicalevaluation or radiological features, improvement of symptomswhile on anti-tuberculosis therapy, and/or on smear positivity;and mixed studies, in which the number of unconfirmed andconfirmed patients was not presented separately, thus possiblyresulting in a selection bias. Three reviewers performed thereview of titles and/or abstracts and 53 publications wereretrieved for full text review (fig. 1).
Data extraction and quality assessmentBoth study selection and data extraction were conductedindependently, in duplicate, by the authors of the presentstudy, in order to reduce the risk of errors. A data extractionform was created, including a subset of predefined itemscovering inclusion and exclusion criteria, as described above.These criteria included: study setting (country); year(s) ofperformance; type of patient and method used to recruitparticipants (including number of eligible and subjects); study
R. DIEL ET AL. TUBERCULOSIS
cEUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 89
duration (from/to); and loss of participants during theobservation period follow-up.
For analysis of specificity, NPV, NPV for progression and PPVfor progression (see section on ‘‘Data synthesis and analysis’’for definitions), the following study contents were documen-ted: IGRA(s) used; mean age or age composition; number ofparticipants finally included; number of subjects for whomvalid test results were available (after subtraction of indeter-minate, i.e. uninterpretable results); duration of follow-up;number of test results considered true negative and falsepositive in persons at very low risk of M. tuberculosis infection(specificity); number of confirmed tuberculosis cases withnegative IGRA scores (i.e. number of false negatives in active
tuberculosis patients to measure NPV); number of confirmednontuberculosis cases with negative IGRA scores (i.e. numbertrue negatives in suspects of active tuberculosis to measureNPV); number of score-negative test results among persons atrisk and screened for LTBI who later developed active disease(i.e. false negatives for NPV for progression); number of score-negative persons screened for LTBI that did not develop activetuberculosis (true negative for NPV for progression calcula-tion); untreated scored-positive persons who developed activetuberculosis during follow-up (PPV for progression); and,lastly, number of scored-positive persons screened for LTBIwith a confirmed relationship to their index case, as demon-strated by highly discriminating molecular genotyping.
For analysis of the correlation between TST and IGRA results,and infection-exposure among BCG-vaccinated or -unvacci-nated subjects, the following items were also included (orrecalculated, if possible, from the available data): time of BCGvaccination; assessment of BCG vaccination among the studyparticipants; TST units; type and chosen cut-off (indurationdiameter); number of participants finally tested by TST and atleast one IGRA; assay used; number of subjects for whom testresults were available; TST and IGRA results (if available, alsoin BCG-vaccinated or -unvaccinated subgroups); number ofindeterminate results (if any); results on IGRAs/IGRA/TSTconcordance and discordance; statistical agreement (k statis-tics) in total and subgroups; and, lastly, odd ratios ofindependent predictors for IGRA and TST positivity. In thecase of TST performance, results using the cut-off chosen bythe authors, as part of the respective study design, were takenfor our review. Additional clarifications on studies wererequested by personal correspondence with the authors. Anydiscrepancies were resolved by consensus with the help of theteam coordinator and, thus, obtaining an inter-revieweragreement of 100%.
Data synthesis and analysisThe following definitions of specificity, NPV, NPV forprogression and PPV for progression were used for analysis.
The specificity was defined as the number of true negativesdivided by the sum of true negatives and false positives. Asthere is no gold standard to ascertain LTBI in a study subject,specificity studies are typically conducted in low-prevalencecountries and settings in which the population is at no risk,and there is no suspicion or evidence of LTBI or priortuberculosis (any individual with a positive test score beingconsidered false-positive). Specificity is, therefore, the measureof a test’s ability to score LTBI-free, healthy subjects as test-negative. This definition holds for both IGRAs and the TST,irrespective of BCG vaccination status.
Generally, the NPV of a screening test is defined as the numberof true negative test results divided by the sum of true andfalse negative results, all in individuals suspected to have adisease. With regards to diagnosing active tuberculosis, thisrefers to the degree to which a test does not score a person withactive tuberculosis as test-negative, i.e. the certainty that ascore-negative person does not have active tuberculosis.Translating this definition to persons suspected of LTBI, theNPV measures the degree to which a test does not score aperson infected with M. tuberculosis as score-negative, i.e. the
432 potentially relevant citations identified by electronic databases and supplementary sources: Specificity (n=123) NPV (n=34) PPV (n=144) TST/IGRA and exposure (n=131)
Excluded before full text screeningSpecificity: Reviews (n=9) Animal study (n=1) Cost analysis (n=7) Guideline (n=1)PPV: No TB investigated (n=42) Reviews (n=9) Animal TB infection (n=4) Cost analysis (n=1)TST/IGRA and exposure: Editorial (n=1) Cost analyses (n=2) Animal TB infection (n=6) Reviews (n=14) Guideline (n=1)
Excluded before in-depth screeningSpecificity: Prior TB exposure/no low-burden country (n=64) No commercial IGRA (n=29) Laboratory studies (n=4) Other reasons (n=4)NPV: No commercial IGRA (n=9) No differentiation between latent and active TB (n=3) Other reasons (n=4)PPV: Laboratory studies (n=7) No data on progression (n=71) Letters without original data (n=3) Other reasons (n=3)TST/IGRA and TB exposure: Laboratory studies (n=7) Experimental study (n=1) No commercial IGRA (n=1) Only active TB cases (n=2) No differentiation between latent and active TB (n=3) Other reasons (n=6)
334 full-text studies screened for further
eligibility criteria
60# studies finally included into review: Specificity (n=4) NPV (n=18) PPV (n=34) TST/IGRAs and TB exposure (n=34)
FIGURE 1. Flow diagram for study selection. NPV: negative predictive value;
assay; TB: tuberculosis. #: seven studies also belonging to another category.
TUBERCULOSIS R. DIEL ET AL.
90 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
certainty that a score-negative person does not have LTBI.Again, due to the lack of a gold standard for LTBI identifica-tion, studies commonly determine the NPV among patientswith confirmed active tuberculosis, using the proportion ofpatients with false-negative test scores as a surrogate for theproportion of false-negative scores in LTBI suspects. Thenumber of true negative cases is consequently the number ofindividuals, suspected of active tuberculosis that are subse-quently confirmed not to have active disease.
NPV for progression is defined as the proportion of IGRAscore-negative subjects that do not progress to active tubercu-losis in a longitudinal, follow-up study of individuals testedfor LTBI. Subsequently, this value reflects a test’s ability tocorrectly predict that an LTBI score-negative individual willnot develop active tuberculosis in later life, provided they haveno further exposure to infection. As only subjects infected withM. tuberculosis can develop disease, the NPV can be measuredby following subjects that scored negative for LTBI over time,and quantifying the number that have remained free fromactive disease.
The PPV for progression is defined as the proportion of IGRAscore-positive subjects that develop active tuberculosis in alongitudinal, follow-up study of individuals tested for LTBI.Subsequently, this value reflects a test’s ability to correctlypredict that an LTBI score-positive individual is at risk ofdeveloping active tuberculosis in later life. As for NPVprogression studies, this value is measured by followingsubjects that scored positive for LTBI over time, and quantify-ing the number that did develop active disease.
Estimates of specificity, NPV and PPV, and their respective95% confidence intervals were calculated for each of theincluded studies and used, if applicable, to calculate pooledestimates. Pooled estimates are weighted averages built byusing the study sample sizes as implicit weights. Weightedpooled estimates were compared by the Pearson’s Chi-squaredtest (or Fisher’s exact test, when the expected cell sizes weresmaller than five). All p-values reported are based on two-tailed comparisons with the statistical significance set atp,0.05.
Statistical heterogeneity between studies was assessed usingthe inconsistency (I2) statistic described by HIGGINS et al. [18].The test describes the percentage of total variability within aset of effect sizes that is caused by true heterogeneity ratherthan by sampling error (chance). I25100%6(Q-df)/Q, where Qis the Chi-squared statistic and df enumerates the degrees offreedom. If applicable, forest plots were built to investigatewhether outcomes were stable over a series of studies.Statistical analyses were performed using Meta-DiSc software,version 1.4 [25], and by applying SPSS version 18.0 forWindows (SPSS Inc., Chicago, IL, USA).
RESULTSThe study selection process is shown in figure 1. 432 citationswere identified, of which 60 articles were eligible for finalinclusion for one or more of the following topics: specificity,NPV in patients with active tuberculosis, NPV for progression,PPV for progression, and association of IGRAs and the TSTwith M. tuberculosis exposure and BCG vaccination. Overall,the quality, evaluated by the QUADAS tool, was very high and
11 (91.7%) out of 12 study quality indicators were met by 100%of the included studies, thereby increasing the strength ofscientific evidence of the review.
SpecificityOnly four published studies evaluated the QFT-G-IT and/or T-SPOT1.TB specificity in a total of 346 subjects with valid IGRAresults [26–29], fulfilling the inclusion criteria. The specificityof the IGRAs ranged from 98% (95% CI 86.8–99.9%) [26] for theT-SPOT1.TB to 100% (95% CI 97.6–100%) [29] for the QFT-G-IT, for which specificity estimates were available in all fourstudies of 99.4% (95% CI 97.9–99.9%) (table 1). We excludedstudies from analysis that either used criteria from the USA forscoring T-SPOT1.TB results [30], or that were performed inintermediate tuberculosis burden countries. In the latter, evenindividuals belonging to low-risk groups may have beenexposed to unknown M. tuberculosis infection with a poten-tially higher risk of infection than that expected in low-burdencountries [31–33]. Regarding the comparison of IGRAs withTST, one report [28] was excluded because the TST was notperformed in all the enrolled subjects. From the data that couldbe analysed, TST specificity ranged from 55% (95% CI 38.5–70.7) [26] to 95% (95% CI 87.7–97.2%) [29], with a pooledspecificity of 88.7% (95% CI 84.6–92.0). When comparing theIGRAs and TST results, only a few subjects not exposed to M.tuberculosis scored positive by IGRAs compared to TST,particularly individuals with known prior BCG vaccinationor confirmed NTM infection (table 1).
NPV in patients with active tuberculosis18 articles satisfied our inclusion criteria for evaluating theNPV; 13 assessed the NPV among confirmed tuberculosiscases [22, 26, 34–44] and six assessed the prospective outcomein IGRA-negative individuals after an average of 2 yrs [19–22,45, 46].
Among the studies in which the NPV was evaluated inpatients with confirmed tuberculosis (387 tuberculosis caseswith a valid T-SPOT1.TB result and 304 with valid QFT-G-ITresult), the NPV varied greatly, irrespective of the IGRA used.Among the 13 articles, the NPV ranged between 74.4% [36] and100% [22, 34], with a pooled value of 94% (95% CI 92.1–95.6%)for the T-SPOT1.TB and 88% (95% CI 84.6–91.5%) for theQFT-G-IT (figs 2 and 3).
In the six longitudinal studies assessing the NPV amonghealthy persons, a total of 1,442 individuals scored negative bythe QFT-G-IT and 182 scored negative by the T-SPOT1.TB(figs 4 and 5) [19–22, 45, 46].
NPV for progressionAlthough subjects at increased risk for developing tuberculosiswere included in all studies, the pooled NPV for progressionwas high in the five studies performed in low-incidencecountries. For these five studies, the pooled NPV for theQFT-G-IT was 99.8% (95% CI 99.4–100%; three individualscontracted tuberculosis among the 1,442 scored-negatives) and97.8% (95% CI 94.5–99.4%) for the T-SPOT1.TB (four indivi-duals contracted active tuberculosis among the 182 scored-negatives). Conversely, the NPV of the included studyperformed in Thailand [46], an intermediate-burden country,was 88% (95% CI 63.6–98.5%).
R. DIEL ET AL. TUBERCULOSIS
cEUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 91
In the majority of the evaluated studies, it was not possible tocompare the NPV estimates for IGRAs to that of the TST results[19, 21, 22, 45, 46]. In particular, in two studies performed onHIV-infected persons, TST-negative individuals were excludedfrom the follow-up [19, 22]. Only one study performed acomprehensive follow-up that included TST and IGRA scored-negative subjects [20]. In this study, only one out of 354 TSTscore-negative close contacts developed tuberculosis, resultingin an NPV for the TST of 99.7% (95% CI 98.5–100%) comparedto 100% for the QFT-G-IT (95% CI 99.4–100%).
PPV for progressionFour studies investigating the probability of developing activedisease in subjects with a positive IGRA result were included foranalysis (table 2). Two studies were screening studies amongHIV-positive subjects, while the other two consisted of largecontact investigations among native and immigrant subjects inGermany and the Netherlands. Only one study [20] includedchildren aged ,16 yrs. The rate of progression to active diseaseamong subjects having tested positive for LTBI, and havingrefused preventive treatment, ranged 2.3–3.3% for TST, 2.8–14.3% for QFT-G-IT and 3.3–10% for T-SPOT1.TB (table 2).
In a study by CLARK et al. [22], 201 HIV-infected patientsdiagnosed with LTBI were followed for a median period of12 months. Two out of the 20 HIV-infected patients withpositive T-SPOT1.TB results, and who did not receivepreventive treatment, developed active tuberculosis, resultingin a PPV for progression of 10% (3 and 10 months after testing,no detail or comparative results with the TST were reported).
In an extensive contact investigation study, DIEL et al. [20]followed the outcome of 41 QFT-G-IT-positive close contacts,all of whom did not receive preventive therapy, for a period off2 yrs. Four of the individuals were TST-negative and threewere children ,16 yrs of age. Within this period, six out of the41 untreated, QFT-G-IT-positive subjects developed activetuberculosis (PPV for progression 14.6%), including oneQFT-G-IT-positive, TST-negative contact. In comparison, ofthe 219 untreated TST-positive contacts followed, only fivedeveloped tuberculosis (PPV for progression 2.3%). Theprogression rate increased to five (5.6%) out of 90 when a10 mm TST cut-off was applied.
AICHELBURG et al. [19] enrolled 830 HIV-infected individuals inan extensive LTBI screening study. Among the study group, 36were QFT-G-IT-positive, subsequently diagnosed with LTBIand followed for a mean period of 19 months withoutreceiving preventive treatment. An additional seven individ-uals also tested positive at the time of screening, but were thendiagnosed with active tuberculosis and subsequently excludedfrom the follow-up. Of the 36 QFT-G-IT-positive individuals,three had developed active tuberculosis within the follow-upperiod, resulting in a PPV for progression of 8.3%. Comparisonbetween TST and QFT-G-IT results with respect to progressionof disease was not possible, due to the fact that TST was notadministered in all individuals.
In their study, KIK et al. [21] included 433 close immigrant,adult contacts in the Netherlands, none of whom wasimmunosuppressed. QFT-G-IT or T-SPOT1.TB was performedonly in TST-positive patients or in persons with a known priorTST result o10 mm (n517). All TST-positive contacts were
TA
BL
E1
Sp
eci
ficity
for
late
nt
Myc
ob
acte
rium
tub
erc
ulo
sis
infe
ctio
n(L
TB
I)d
iag
no
sis
inin
div
idu
als
with
very
low
risk
for
LTB
Ite
sted
by
Qu
an
tiFE
RO
N1
-Go
ldIn
-Tu
be
(QF
T-G
-IT)
an
d/o
rT-S
PO
T1
.TB
ass
ays
Fir
st
au
tho
r
[re
f.]
Co
un
try
(bu
rde
n)
Pa
rtic
ipa
nts
IGR
A(s
)u
se
dA
ge
me
an¡
SD
#
yrs
Su
bje
cts
assig
ne
da
s
low
risk
for
M.
tub
erc
ulo
sis
infe
cti
on
n
Su
bje
cts
wit
hn
eg
ati
ve
tests
/all
teste
dsu
bje
cts
"
n/N
Su
bje
cts
co
nsid
ere
d
fals
e-p
osit
ive
n
Ro
un
de
dsp
ecif
icit
y
of
IGR
A(s
)/T
ST
%
(95
%C
I)
DE
TJE
N[2
6]
Germ
an
y(lo
w)
Ch
ildre
nw
ithco
nfir
med
no
n-T
B
lym
ph
ad
en
itis
(med
ian
44)
or
resp
irato
ryin
fect
ion
s
T-S
PO
T1
.TB
/QF
T-G
-IT
NTM
infe
ctio
n:
med
ian
44
mo
nth
s;o
thers
:
52.5
mo
nth
s
45
T-S
PO
T1.T
B:
39/4
0;Q
FT-
G-I
T:
40/4
0;TS
T:
22/4
0
T-S
PO
T1
.TB
:1;
QF
T-G
-IT:
0;
TS
T:
18
(all
18
with
NTM
lym
phad
eniti
s)
T-S
PO
T1
.TB
:98
(86.8
–99.9
);
QF
T-G
-IT:
100
(92.8
–100);
TS
T:
55(3
8.5–
70.7
)
FR
AN
KE
N[2
7]
Th
eN
eth
erla
nd
s(lo
w)
Du
tch
Arm
ed
Fo
rces
pers
on
nel
QF
T-G
-IT
19.6
¡2.8
171
QF
T-G
-IT:
166/1
68
+ ;TS
T:
136/
145
QF
T-G
-IT:
2;
TS
T:
9Q
FT-G
-IT:
99
(95.8
–99.9
);TS
T:
94
(88.6
–97.1
)
PA
LA
ZZ
O[2
8]
Italy
(low
)H
ealth
yb
loo
dd
on
ors
as
con
tro
ls
for
TB
susp
ect
s
QF
T-G
-IT
BC
G-u
nva
cc.:
37.2
¡2;
BC
G-v
acc
.:35.2
¡2
24
QF
T-G
-IT:
14/1
41;
TS
T:
NA
QF
T-G
-IT:
0Q
FT-
G-I
T:
100
(80.
7–1
00);
TS
T:
NA
RU
HW
AL
D[2
9]
Den
mark
(low
)86
hig
hsc
ho
ol
stu
den
tsan
d38
hig
hsc
ho
ol
staff
QF
T-G
-IT
Stu
dents
:17.6
¡1.3
;st
aff
:
54.5
¡8.5
124
QF
T-G
-IT:1
24/1
24;TS
T:
116/1
24
QF
T-G
-IT:
0;
TS
T:
8Q
FT-
G-I
T:
100
(97.
6–1
00);
TS
T:
95
(87.7
–97.
2)
Po
ole
dsp
eci
ficity
ofQ
FT-G
-IT:9
9.4
%(9
5%
CI9
7.9
–99.
9%
;C
hi-s
quare
d(Q
)52.
90;d
eg
rees
off
reed
om
(df)
53;p
50.4
069;i
nco
nsi
sten
cy(I
2)5
0.0
%.P
oo
led
speci
ficity
oft
ub
erc
ulin
skin
test
(TS
T):
88.7
%(9
5%
CI8
4.6
–92.0
%;
Q5
2.90;
df5
2;
p5
0.0001;
I25
94.5
%).
IGR
A:
inte
rfer
on
-cre
lease
assa
y;TB
:tu
berc
ulo
sis;
NTM
:n
on
tub
erc
ulo
us
myc
ob
act
eria
;B
CG
:b
aci
lleC
alm
ette–G
uer
in;
unva
cc.:
unva
ccin
ate
d;
vacc
.:va
ccin
ate
d;
NA
:n
ot
ass
ess
ed
.#
:u
nle
sso
therw
ise
state
d;
":aftersu
btr
act
ion
ofi
nva
lidre
sults
/ind
ete
rmin
ate
s;+ :
amo
ng
recr
uits
,fiv
eo
uto
f171
resu
ltsw
ere
QF
T-G
-IT-p
osi
tive;o
fth
ese
recr
uits
,tw
oh
ad
pre
vio
usl
yb
een
trea
ted
fortu
berc
ulo
sis
an
do
ne
was
fore
ign-b
orn
with
ou
tin
form
atio
no
nth
etu
berc
ulo
sis
pre
vale
nce
inh
ish
om
eco
untr
y,th
us
red
uci
ng
the
num
ber
of
fals
e-p
osi
tive
sub
ject
san
dth
en
um
ber
of
test
ed
pers
ons
elig
ible
for
an
aly
sis
acc
ord
ing
ly;
1:
13
BC
G-
un
vacc
inat
ed
an
do
ne
-vacc
inate
dco
ntr
olp
ers
ons,
as
sho
wn
inta
ble
2o
fth
ep
ub
licatio
n;
evalu
atio
no
fTS
Tre
sults
was
no
tp
erf
orm
ed
beca
use
TS
Tw
as
perf
om
ed
ino
nly
seve
no
uto
fth
e14
pers
ons
incl
ud
ed
inth
est
ud
y.
TUBERCULOSIS R. DIEL ET AL.
92 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
followed for f24 months. In total, 339 (78.3%) of the 433contacts were TST-positive. IGRA was performed on 327 of theTST-positive contacts, of which 179 (54.4%) tested positiveusing the QFT-G-IT and 181 (60.5%) tested T-SPOT1.TBpositive. Nine (2.6%) of the 339 contacts developed activetuberculosis; IGRA had been performed in eight of thesecontacts, of whom six had scored positive. The PPV value fordeveloping active tuberculosis was 3.1% (nine out of 288; 95%CI 1.4–5.8%) for TST (cut-off o10 mm), 2.8% (five out of 178;95% CI 0.9–6.4%) for QFT-G-IT and 3.3% (six out of 181; 95% CI1.2–7.0%) for T-SPOT1.TB.
Association of IGRAs and the TST with MTB exposure andBCG vaccination34 studies [27, 39, 45, 47–77] from a total of 18 differentcountries fulfilled the inclusion criteria for this section of the
review. Four studies were from intermediate-burden countries(South Korea [55], Lithuania [59], Iran [63] and Turkey [72])and three studies were from high-burden countries (India [56],South Africa [61] and Vietnam [65]). Seven studies were basedsolely on paediatric cohorts [39, 51, 54, 59, 73, 75] and threestudies [47, 57, 61] included children in the study population.
17 of the 34 articles calculated the odd ratios, assessing theinfluence of different factors on test positivity in multivariateanalyses. In nine of these, TST positivity was significantlyassociated with BCG vaccination as an independent predictorin multivariate analysis [47, 50, 55, 62, 67, 71, 72, 74, 77] withodd ratios ranging between 3.8 (95% CI 1.0–13.9) [50] and 24.7(95% CI 11.7–52.5) [48] regardless of the epidemiologicalcontext. No correlation was found between IGRA positivityand BCG vaccination. In nine of the 10 studies comparing the
Pooled NPV: 0.88 (0.85_0.92)Chi-squared=40.32; df=6 (p=0.0000)Inconsistency I2=85.1%
NPV 95% CI Patients n/N
FIGURE 3. Negative predictive value (NPV) of QuantiFERON1-TB Gold In-Tube assay in tuberculosis suspects. df: degrees of freedom.
R. DIEL ET AL. TUBERCULOSIS
cEUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 93
odds for test positivity with M. tuberculosis exposure gradients,or using chest radiography lesions as a surrogate of priorexposure, the IGRAs associated better with exposure than theTST, irrespective of the setting’s disease burden. Furthermore,there was generally a poor agreement between IGRAs and TSTresults (for k statistics, see table 3).
DISCUSSIONIn the past few years, a large number of studies have beenpublished on the performance of IGRAs for the diagnosis ofLTBI in different cohorts at risk for progression to activetuberculosis. We performed a systematic review and meta-analysis to assess the accuracy of commercially availableIGRAs in terms of specificity, NPV and PPV as the mainestimates. Despite the volume of publications on the subject,only a limited number of studies fulfilled the inclusion criteriafor our systematic review and meta-analysis. This is due to theuse of noncommercial IGRAs or methodological limits in thedesign of individual studies.
Based on the results obtained from our analysis we identifiedthat: 1) the IGRAs show a higher specificity compared with theTST; 2) the NPV in patients with active tuberculosis as asurrogate of LTBI is high (although with a high variability invalues); 3) the ability of the IGRAs to predict that a test-negative individual will not develop disease is even higher;and 4) the predictive value for progression to active disease
when testing positive is higher for the IGRAs compared withthe TST. However, NPV and PPV remain to be established inthe paediatric population and in immunocompromised indi-viduals, e.g. patients with advanced HIV infection, transplantrecipients or candidates for TNF antagonists therapies.
The pooled specificity (99.4%, 95% CI 97.8–99.9%) whenmeasured in low-risk population groups using the QFT-G-IT(the only IGRA for which results could be analysed in allincluded studies) was clearly higher than the pooled specificityfor the TST (88.7%, 95% CI 84.6–92.0%). These results suggestthat the IGRAs are more certain to correctly identifyindividuals not infected with M. tuberculosis as compared tothe TST. However, the estimates have to be interpreted withcaution due to the low number of included studies.
An optimal LTBI diagnostic tool should be predictive of anindividual’s risks for developing or not developing activedisease when tested. When analysing the studies using cohortsof active tuberculosis cases as surrogates of LTBI, a high overallNPV value was measured (pooled NPV 94% for T-SPOT1.TBand 88% for QFT-G-IT). This would suggest that IGRAs,especially the T-SPOT1.TB, are effective at ruling out M.tuberculosis infection. However, caution must be exercisedwhen interpreting these results, as there was a high variabilitybetween the studies (ranging 70–98% for the QFT-G-IT and 92–100% for the T-SPOT1.TB), probably due to the corresponding
Pooled NPV: 0.998 (0.994_1.0)Chi-squared=13.67; df=3 (p=0.0034)Inconsistency I2=78.1%
FIGURE 4. Negative predictive value (NPV) for progression in QuantiFERON1-TB Gold In-Tube assay negative subjects. df: degrees of freedom.
NPV 95% CI
1.00 0.92–1.000.98 0.94–1.000.88 0.64–0.99
Patients n/N
47/47116/11815/17
CLARK [22]KIK [21]LEE [41]
0 0.2 0.4NPV
0.6 0.8 1
■
■■
■
Pooled NPV: 0.98 (0.94_0.99)Chi-squared=5.86; df=2 (p=0.0533)Inconsistency I2=65.9%
FIGURE 5. Negative predictive value (NPV) for progression in T-SPOT1.TB negative subjects. df: degrees of freedom.
TUBERCULOSIS R. DIEL ET AL.
94 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
prevalence of active tuberculosis cases, thus limiting the utilityof pooling data across studies. This would, therefore, refute theconclusion that the IGRAs are an effective rule-out test for LTBI.
Conversely, longitudinal NPV studies on mostly immuno-competent adults in low-prevalence countries, who are at riskfor LTBI, suggest that one can be reasonably confident that,when scoring negative with an IGRA, the likelihood of falsenegative results is low. High NPVs were found for both theQFT-G-IT and the T-SPOT1.TB (pooled NPV of 97.8% for T-SPOT1.TB and 99.8% for QFT-G-IT). This indicates that anindividual testing negative will most likely not developtuberculosis in the future. However, a limitation of thisanalysis was the low number of individuals included in thestudies (n51,442), the short durations of follow-up and the factthat similar studies have not yet been performed in high-burden settings.
The QFT-G-IT and the T-SPOT1.TB both showed a similar PPVfor progression, with a slightly wider range in values ascompared to the TST (2.8–14.3% for QFT-G-IT; 3.3–10% for T-SPOT1.TB and 2.3–3.3% for TST), suggesting that the IGRAshave a higher predictive value for progression to active diseaseas compared with the TST. If IGRA performance was notconditionally restricted to prior TST positivity (as was the casein one study), the progression rates were higher (8, 10 and15%) than those reported for TST, with 2 yrs follow-up [19, 20,22]. In the study by KIK et al. [21], a lower PPV value forprogression was measured (3%); however, the study waslimited to assessing the IGRA among TST-positive individuals,thereby presenting a source of bias.
Studies on predictive value of the IGRAs with regard todeveloping active tuberculosis upon testing positive are stillvery few, with only four studies having fulfilled our inclusioncriteria, and they often vary substantially in design and rely onempirical observations of subjects refusing LTBI treatment.Additional, larger sized studies including new markers forLTBI are needed to evaluate the predictive values of IGRAsand other biomarkers in groups with the highest risk ofprogression to tuberculosis, especially in children and inimmunocompromised hosts.
The review of studies investigating the influence of BCGvaccination on TST positivity and the influence of exposuresfor LTBI clearly confirmed, despite the large heterogeneity inthe design of the analysed studies, that, unlike the TST, thenewest commercial IGRAs are not affected by prior BCGvaccination. They are subsequently more likely to be associatedwith exposure to tuberculosis cases compared to the TST.These results were not influenced by the epidemiologicalcontext where the studies were performed (high- versus low-incidence countries).
The review only assessed TST results in the context of IGRAstudies and, thus, the results on prior studies on the accuracyperforming the TST alone could not be taken into consideration.
Furthermore, we did not evaluate the ability of IGRAs todiscriminate recent from remote LTBI and the phenomenon ofconversion/reversion over time or after therapy. Although thiswould be a very important point in a public healthprogramme, the data available today [78–82] are limited and
TA
BL
E2
Po
sitiv
ep
red
ictiv
eva
lue
(PP
V)
of
com
merc
ial
inte
rfero
n-c
rele
ase
ass
ays
(IG
RA
s)fo
rp
rog
ress
ion
inth
ose
with
late
nt
Myc
ob
acte
rium
tub
erc
ulo
sis
infe
ctio
n
Fir
st
au
tho
r
[re
f.]
Stu
dy
de
sig
n
Pa
rtic
ipa
nts
Co
un
try
(bu
rde
n)
IGR
A(s
)u
se
dA
ge
me
an¡
SD
#
yrs
Pe
rfo
rma
nce
of
IGR
A
an
dT
ST
testi
ng
Fo
llo
w-u
p
pe
rio
dm
on
ths
La
ter
TB
ca
se
sa
mo
ng
un
tre
ate
dIG
RA
/TS
T
po
sit
ive
sn
/N(P
PV
%
(95
%C
I))"
Late
rT
Bca
ses
am
on
gIG
RA
/
TS
T-p
osi
tive
sw
ith
co
nfirm
ed
rela
tio
nsh
ipto
ind
ex
case
s
by
RF
LP
n/N
(%)
CL
AR
K[2
2]
Case
–
con
tro
l
201
imm
un
e-d
efic
ien
t
HIV
-1-p
osi
tive
ad
ults
UK
(low
)T-S
PO
T1
.TB
Med
ian
40
(ran
ge
33–4
6)
NA
24
T-S
PO
T1
.TB
:2/2
0(1
0(0
.012–0
.32))
;
TS
T:
NA
NA
DIE
L[2
0]
Co
ho
rt601
clo
seco
nta
cts
of
AF
B-p
osi
tive
TB
case
s
(66
(10.9
%)
child
ren
,16
yrs)
Germ
an
y(lo
w)
QF
T-G
-IT
27.7
¡12.0
(ran
ge
1–5
6)
Sim
ulta
neo
usl
y
with
TS
T
24
(mean
)Q
FT-G
-IT:
6/4
1(1
4.6
(0.0
6–0
.29))
;5/2
19
TS
T-p
osi
tives
.5
mm
:(2
.3(0
.007–
0.0
52))
QF
T-G
-IT:
2/6
(33.3
);
TS
T:
1/5
(20)
AIC
HE
LB
UR
G[1
9]
Co
ho
rt830
HIV
-1-p
osi
tive
ad
ults
Au
stria
(low
)Q
FT-G
-IT
39
(inte
rqu
art
ile
ran
ge
32–4
7)
IGR
Afir
st,
TS
To
nly
if
QF
T-G
-IT
po
sitiv
e
19
(mean
)Q
FT-G
-IT:
3/3
6+
(8.3
(0.0
18–0
.22))
;TS
T:
NA
NA
KIK
[21
]C
oh
ort
433
ad
ult
clo
se
imm
igra
nt
con
tact
san
d
BC
G-v
acc
.D
utc
h-b
orn
con
tact
s
Th
eN
eth
erla
nd
s
(low
)
QF
T-G
-IT,
T-S
PO
T1
.TB
NA
TS
Tfir
st,
IGR
Ao
nly
if
TS
Tp
osi
tive
(at
a
cut-
offo
5m
m)1
22
e(m
ed
ian
)Q
FT-G
-IT:
5/1
78
(2.8
(0.0
09–0
.064))
;
T-S
PO
T1.T
B:6
/181
(3.3
(0.0
12–0
.07)
);TS
T
o10
mm
:9/
288
(3.1
(0.0
14–0
.058
))#
#
IGR
As:
3/3
(100)"
";
TS
T:
6/6
(100)
TS
T:
tub
erc
ulin
skin
test
;TB
:tu
berc
ulo
sis;
RF
LP:
rest
rictio
nfr
ag
men
tle
ng
thp
oly
mo
rph
ism
;N
A:
no
tass
ess
ed
;A
FB
:aci
d-f
ast
baci
lli;
QF
T-G
-IT:
Qu
an
tiFE
RO
N1
-TB
Go
ldIn
-Tu
be
ass
ay;
BC
G:
baci
lleC
alm
ette–G
uerin
;va
cc.:
vacc
inate
d.
#:
un
less
oth
erw
ise
state
d;
":
aft
er
sub
tract
ion
of
inva
lidre
sults
/ind
ete
rmin
ate
s;+ :
am
on
g44
QF
T-G
-IT-p
osi
tive
patie
nts
,se
ven
were
dia
gno
sed
with
act
ive
TB
du
ring
the
first
inve
stig
atio
nan
do
ne
was
lost
with
inth
efir
st3
weeks
of
follo
w-u
p;
1:
con
ven
tion
alw
aiti
ng
perio
do
f
8w
eeks
was
no
tresp
ect
ed
for
IGR
As
or
the
TS
T;TS
Tw
as
rep
eate
d8–1
2w
eeks
late
rif
neg
ativ
ein
itially
,bu
tno
tth
eIG
RA
ifth
eIG
RA
resu
ltw
as
initi
ally
neg
ativ
e;
e:
on
ly95
(53.4
%)
ou
tof1
78
oft
he
QF
T-G
-IT-p
osi
tive
con
tact
san
do
nly
101
(55.8
%)
ou
tof1
81
of
the
T-S
PO
T1
.TB
-po
sitiv
eco
nta
cts
were
follo
wed
for
21
mo
nth
s;#
#:
as
calc
ula
ted
by
the
au
tho
rso
fth
at
stu
dy,
alth
oug
ha
TS
Tw
as
con
sid
ere
dp
osi
tive
ifin
du
ratio
nd
iam
ete
rw
aso
5m
m;
"":
six
ou
to
fn
ine
patie
nts
deve
lop
ing
tub
erc
ulo
sis
were
cultu
re-
con
firm
ed
;tw
oo
fth
em
were
neg
ativ
eb
yb
oth
IGR
As
an
d,
ino
ne
tub
erc
ulo
sis
patie
nt,
cultu
re-c
on
firm
ed
IGR
Ate
stin
gw
as
no
tp
erf
orm
ed
.
R. DIEL ET AL. TUBERCULOSIS
cEUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 95
TA
BL
E3
Stu
die
sin
wh
ich
exp
osu
rest
atu
sw
as
ap
red
icto
ro
fte
stp
osi
tivity
(eith
er
test
)in
mu
ltiva
riate
an
aly
sis
Fir
st
au
tho
r[r
ef.
]C
ou
ntr
y
(IG
RA
use
d)
Pa
rtic
ipa
nts
BC
Gva
ccin
ee
s
n/N
(%)
IGR
Ap
osit
ive
s/
tota
lte
ste
dn
/N
IGR
AO
R(9
5%
CI)
TS
Tp
osit
ive
s/t
ota
l
teste
dn
/N(%
)
TS
TO
R(9
5%
CI)
Sta
tisti
ca
la
gre
em
en
t
be
twe
en
IGR
A(s
)a
nd
TS
Tk
(95
%C
I#)
BA
RT
AL
ES
I[5
0]
Italy
(QF
T-G
-IT)
LTB
Isc
reen
ing
in398
IMID
sub
ject
s
16/3
93
(4.1
)52/3
98
5.3
(1–2
8)
75/3
23
(23.2
)6.5
(1.1
–36.9
)0.5
5"
(0.4
4–0
.66)
CA
SA
S[5
3]
Sp
ain
(T-S
PO
T1
.TB
an
d
QF
T-G
-IT)
147
HC
Ws
23/1
47
(15.6
)T-S
PO
T.T
B:
57/1
57;
QF
T-G
-IT:
43/1
47
T-S
PO
T:
2.1
(1.5
–4.1
);
QF
T:
1.8
2(0
.88–3
.8)
103/1
47
(70.1
)0.9
4+
(0.4
6–1
.93)
T-S
PO
T.T
B:
0.3
2(S
E0.0
61),
BC
G:
0.1
7(S
E0.0
9);
QF
T-G
-IT:k
(to
tal):
0.2
9
(SE
0.0
52),
BC
G:
0.0
85
(SE
0.0
5)
HE
SS
EL
ING
[61
]S
ou
thA
fric
a
(T-S
PO
T1
.TB
an
d
QF
T-G
-IT)
82
clo
seco
nta
cts
(29
child
ren
an
d53
ad
ults
)
62/8
2(7
5.6
)29/7
4T-S
PO
T.T
B:
38.4
0
(7.5
9–6
16.1
1);
QF
T-G
-IT:
14.9
4
(4.0
2–5
5.5
8)
54/7
8(6
9.2
)3.8
3(1
.05–1
4.0
3)
T-S
PO
T.T
B:
0.1
2"
(0.1
1–0
.36);
QF
T-G
-IT:
0.4
5"
(0.2
8–0
.62)
LAF
FIT
TE
[64
]S
witz
erla
nd
(T-S
PO
T1
.TB
)
LTB
Isc
reen
ing
in50
pso
riasi
sp
atie
nts
befo
rean
ti-
TN
F-a
thera
py
46/5
0(9
2.8
)10/5
0C
on
tact
his
tory
:5.6
7
(1.2
5–2
5.7
);le
sio
ns
on
CX
R:
25.3
(2.4
1–2
67)
20/5
0(4
0.0
)C
uto
ff.
5m
m:
2.1
4+
(0.5
5–8
.3);
.10
mm
:1.6
7+
(0.4
3–6
.5)
0.3
3"
(SD
0.1
3)
LIE
N[6
5]
Vie
tnam
(QF
T-G
-IT)
300
HC
Ws
Less
than
on
e-t
hird
(data
no
tsh
ow
n)
142/3
00
1.9
4(1
.04–3
.64)
176/2
88
(61.1
)N
ot
calc
ula
ted
0.4
4(S
E0.0
6),
BC
G:
0.2
9( S
E0.0
9)
MA
TU
LIS
[67
]S
witz
erla
nd
(QF
T-G
-IT)
LTB
Isc
reen
ing
in142
rheu
matic
patie
nts
118/1
42
(83.1
)17/1
42
Co
nta
cth
isto
ry:
17.8
(2.0
6–1
54);
lesi
on
sin
CX
R:
66.8
(10.1
–441);
46/1
15
(40.0
)6.2
3(1
.18–3
3.0
1)
0.1
7"
(0.0
2–0
.32)
SE
YH
AN
[72
]Tu
rkey
(QF
T-G
-IT)
LTB
Isc
reen
ing
in100
haem
od
ialy
sis
patie
nts
72/1
00
(72.0
)43/1
00
Co
nta
cth
isto
ry:
5.0
8
(1.2
–21.2
);
lesi
on
sin
CX
R:
3.0
6(2
.1–1
1.9
)
34/1
00
(34.0
)N
ot
calc
ula
ted
0.2
6,
BC
G:
0.1
5
TR
IVE
RIO
[74
]S
witz
erla
nd
(T-S
PO
T1
.TB
an
dQ
FT-G
-IT)
LTB
Isc
reen
ing
in62
ES
RD
patie
nts
14/6
2(2
2.6
)T-S
PO
T1
.TB
:18/6
2;
QF
T-G
-IT:
13/6
2
T-S
PO
T1
.TB
:1.2
(0.3
–4.8
;
QF
T-G
-IT:
4.6
(1.2
–18.1
)
12/6
2(1
9.4
)1.4
+(0
.3–7
.0)
QF
T-G
-IT:
0.1
6";
T-S
PO
T1
-TB
:0.3
2"
VIN
TO
N[7
7]
Au
stra
lia(Q
FT-G
-IT)
LTB
Isc
reen
ing
in481
ho
spita
lst
aff
mem
bers
fro
m5
ho
spita
ls
375/4
81
(78.0
)32/3
81
5.6
(1.4
2–2
2.0
)120/3
64
(33.0
)1.9
6+
(0.6
8–5
.63)
0.1
6"
ZE
LL
WE
GE
R[4
8]
Sw
itzerla
nd
(T-S
PO
T1
.TB
)
91
con
tact
pers
on
s
(resi
dents
/sta
ffm
em
bers
)
inan
inst
itutio
nfo
r
alc
oh
olic
patie
nts
78/9
1(8
5.7
)15/9
15.0
(1.0
5–2
3.8
6)
40/9
1(4
4.0
)1.8
5+
(0.7
8–4
.36)
0.2
32
"
IGR
A:
inte
rfero
n-c
rele
ase
ass
ay;
BC
G:
baci
lleC
alm
ett
e–G
ue
rin;
TS
T:
tub
erc
ulin
skin
test
;Q
GT-G
-IT:
Qu
an
tiFE
RO
N1
-TB
Go
ldIn
-Tu
be
ass
ay;
LTB
I:la
ten
tM
yco
bac
teriu
mtu
be
rculo
sis
infe
ctio
n;
IMID
:im
mu
no
med
iate
d
infla
mm
ato
ryd
isease
s;H
CW
:h
ealth
care
wo
rker;
TN
F:
tum
ou
rn
ecr
osi
sfa
cto
r;C
XR
:ch
est
rad
iog
rap
hy:
ES
RD
:en
d-s
tag
ere
nal
dis
ease
.#
:u
nle
sso
therw
ise
state
d;
":
no
div
isio
nin
tosu
bg
rou
ps;
+ :n
ot
statis
tically
sig
nifi
can
t.
TUBERCULOSIS R. DIEL ET AL.
96 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
mostly obtained with experimental techniques and in parti-cular settings, making it difficult to interpret their potentialclinical value. In conclusion, the present systematic review andmeta-analysis of the accuracy of the IGRAs for LTBI diagnosisconfirm the concept that the IGRAs are a valid alternative tothe TST. The superior specificity and the good NPV make themthe first choice, especially in BCG-vaccinated subjects. Thecurrent evidence is, however, still limited in determiningwhether the IGRAs have a stronger predictive value fordeveloping active disease later in life, as compared to theTST. In the development of guidance on the use of IGRAs intuberculosis screening programmes, it is vital that suchevidence is considered when describing diagnostic algorithmsfor the diagnosis of LTBI. In this way, we can provide supportto assure that only assays with proven accuracy are introducedinto national screening programmes for further advancementtowards the elimination of tuberculosis.
STATEMENT OF INTERESTStatements of interest for D. Goletti, G. Ferrara, B. Kampmann,M. Ruhwald and D. Wagner are available from www.erj.ersjournals.com/site/misc/statements.xhtml
ACKNOWLEDGEMENTSThe authors’ affiliations are as follows: R. Diel (Dept of Pneumology,Medical School Hanover, Hanover, Germany); D. Goletti (Dept ofEpidemiology and Preclinical Research, National Institute forInfectious Diseases L. Spallanzani, Rome, Italy); G. Ferrara (Sectionof Respiratory Diseases, Dept of Internal Medicine, S Maria Hospital,University of Perugia, Terni, Italy); G. Bothamley (HomertonUniversity Hospital, NHS Foundation Trust, London, UK); D. Cirillo(Emerging Bacterial Pathogens Unit, San Raffaele Scientific Institute,Milan, Italy); B. Kampmann (Academic Dept of Paediatrics, ImperialCollege, London); C. Lange (Division of Clinical Infectious Diseases,Research Center Borstel, Borstel, Germany); M. Losi (Dept of Oncology,Hematology and Respiratory diseases, University of Modena andReggio Emilia, Modena, Italy); R. Markova (Immunology andAllergology, National Center of Infectious and Parasitic Diseases,Sofia, Bulgaria); G.B. Migliori (WHO Collaborating Centre forTuberculosis and Lung Diseases, Tradate, Italy); A. Nienhaus(Institute for Health Service Research in Dermatology and Nursing,University Medical Center Hamburg-Eppendorf, Hamburg, Germany);M. Ruhwald (Clinical Research Centre, Hvidovre Hospital, Copen-hagen University, Copenhagen, Denmark); D. Wagner (Division ofInfectious Diseases and Centre of Chronic Immunodeficiency, Univer-sity Medical Center, Freiburg, Germany); J.P. Zellweger (Dept ofAmbulatory Care and Community Medicine, University of Lausanne,Lausanne, Switzerland; E. Huitric, A. Sandgren and D. Manissero (bothEuropean Centre for Disease Prevention and Control (ECDC),Stockholm, Sweden).
The authors thank M. Sester (Institute of Transplantation and Infec-tion Immunology, University of Saarland, Homburg, Germany) andC. Giehl (Head of Life Science Division, EURICE, Saarbrucken,Germany) for management and editorial support, and G. Sotgiu(Institute for Biostatistics and Epidemiology, University Sassari,Sassari, Italy) for statistical advice. The Tuberculosis Network Euro-pean Trials Group (TBNET) is a Clinical Research Collaboration (CRC)of the European Respiratory Society (ERS).
REFERENCES1 Raviglione MC, Uplekar MW. WHO’s new Stop TB Strategy.
Lancet 2006; 367: 952–955.
2 European Centre for Disease Prevention and Control. Framework
Action Plan to Fight Tuberculosis in the European Union. www.ecdc.europa.eu/en/publications/Publications/0803_SPR_TB_Action_plan.
pdf Date last accessed: July 7, 2010. Date last updated: 2008.
3 Leung CC, Rieder HL, Lange C, et al. Treatment of latent infectionwith M. tuberculosis: update 2010. Eur Respir J 2010; [Epub ahead of
print DOI: 10.1183/?09031936.00079310].
4 Bothamley GH, Ditiu L, Migliori GB, et al. Active case finding oftuberculosis in Europe: a Tuberculosis Network European Trials
Group (TBNET) survey. Eur Respir J 2008; 32: 1023–1030.
5 Erkens C, Kamphorst M, Abubakar I, et al. Tuberculosis contactinvestigation in low prevalence countries: a European consensus.
Eur Respir J 2010; 36: 925–949.
6 Targeted tuberculin testing and treatment of latent tuberculosisinfection. Am J Respir Crit Care Med 2000; 161: S221–S247.
7 Andersen P, Munk ME, Pollock JM, et al. Specific immune-based
diagnosis of tuberculosis. Lancet 2000; 356: 1099–1104.
19 Aichelburg MC, Rieger A, Breitenecker F, et al. Detection and
prediction of active tuberculosis disease by a whole-bloodinterferon-c release assay in HIV-1-infected individuals. Clin
Infect Dis 2009; 48: 954–962.
20 Diel R, Loddenkemper R, Meywald-Walter K, et al. Predictive
value of a whole blood IFN-c assay for the development of active
tuberculosis disease after recent infection with Mycobacterium
tuberculosis. Am J Respir Crit Care Med 2008; 177: 1164–1170.
21 Kik SV, Franken WP, Mensen M, et al. Predictive value for
progression to tuberculosis by IGRA and TST in immigrantcontacts. Eur Respir J 2010; 35: 1346–1353.
22 Clark SA, Martin SL, Pozniak A, et al. Tuberculosis antigen-
specific immune responses can be detected using enzyme-linkedimmunospot technology in human immunodeficiency virus
(HIV)-1 patients with advanced disease. Clin Exp Immunol 2007;150: 238–244.
R. DIEL ET AL. TUBERCULOSIS
cEUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 97
23 Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items forsystematic reviews and meta-analyses: the PRISMA statement.PLoS Med 2009; 6: e1000097.
24 Whiting P, Rutjes AW, Reitsma JB, et al. The development ofQUADAS: a tool for the quality assessment of studies of diagnosticaccuracy included in systematic reviews. BMC Med Res Methodol
2003; 3: 25.
25 Zamora J, Abraira V, Muriel A, et al. Meta-DiSc: a software formeta-analysis of test accuracy data. BMC Med Res Methodol 2006; 6:31.
26 Detjen AK, Keil T, Roll S, et al. Interferon-c release assays improvethe diagnosis of tuberculosis and nontuberculous mycobacterialdisease in children in a country with a low incidence oftuberculosis. Clin Infect Dis 2007; 45: 322–328.
27 Franken WP, Timmermans JF, Prins C, et al. Comparison ofMantoux and QuantiFERON TB Gold tests for diagnosis of latenttuberculosis infection in Army personnel. Clin Vaccine Immunol2007; 14: 477–480.
28 Palazzo R, Spensieri F, Massari M, et al. Use of whole-bloodsamples in in-house bulk and single-cell antigen-specific gammainterferon assays for surveillance of Mycobacterium tuberculosis
29 Ruhwald M, Petersen J, Kofoed K, et al. Improving T-cell assays forthe diagnosis of latent TB infection: potential of a diagnostic testbased on IP-10. PLoS ONE 2008; 3: e2858.
30 Bienek DR, Chang CK. Evaluation of an interferon-c release assay,T-SPOT.TB, in a population with a low prevalence of tuberculosis.Int J Tuberc Lung Dis 2009; 13: 1416–1421.
31 Harada N, Higuchi K, Yoshiyama T, et al. Comparison of thesensitivity and specificity of two whole blood interferon-c assaysfor M. tuberculosis infection. J Infect 2008; 56: 348–353.
32 Higuchi K, Kawabe Y, Mitarai S, et al. Comparison of performancein two diagnostic methods for tuberculosis infection. Med MicrobiolImmunol 2009; 198: 33–37.
33 Lee JY, Choi HJ, Park IN, et al. Comparison of two commercialinterferon-c assays for diagnosing Mycobacterium tuberculosisinfection. Eur Respir J 2006; 28: 24–30.
34 Dheda K, van Zyl-Smit RN, Meldau R, et al. Quantitative lung Tcell responses aid the rapid diagnosis of pulmonary tuberculosis.Thorax 2009; 64: 847–853.
35 Goletti D, Vincenti D, Carrara S, et al. Selected RD1 peptides foractive tuberculosis diagnosis: comparison of a gamma interferonwhole-blood enzyme-linked immunosorbent assay and anenzyme-linked immunospot assay. Clin Diagn Lab Immunol 2005;12: 1311–1316.
36 Goletti D, Carrara S, Butera O, et al. Accuracy of immunodiagnos-tic tests for active tuberculosis using single and combined results:a multicenter TBNET-Study. PLoS ONE 2008; 3: e3417.
37 Jafari C, Ernst M, Kalsdorf B, et al. Rapid diagnosis of smear-negative tuberculosis by bronchoalveolar lavage enzyme-linkedimmunospot. Am J Respir Crit Care Med 2006; 174: 1048–1054.
38 Jafari C, Thijsen S, Sotgiu G, et al. Bronchoalveolar lavage enzyme-linked immunospot for a rapid diagnosis of tuberculosis: aTuberculosis Network European Trials group study. Am J Respir
Crit Care Med 2009; 180: 666–673.
39 Kampmann B, Whittaker E, Williams A, et al. Interferon-c releaseassays do not identify more children with active tuberculosis thanthe tuberculin skin test. Eur Respir J 2009; 33: 1374–1382.
40 Kang YA, Lee HW, Hwang SS, et al. Usefulness of whole-bloodinterferon-c assay and interferon-c enzyme-linked immunospotassay in the diagnosis of active pulmonary tuberculosis. Chest2007; 132: 959–965.
41 Lee LN, Chou CH, Wang JY, et al. Enzyme-linked immunospotassay for interferon-c in the diagnosis of tuberculous pleurisy. ClinMicrobiol Infect 2009; 15: 173–179.
42 Markova R, Todorova Y, Drenska R, et al. Usefulness ofinterferon-c release assays in the diagnosis of tuberculosis
infection in HIV-infected patients in Bulgaria. Biotechnol
Biotechnol Eq 2009; 23: 1103–1108.
43 Wang JY, Chou CH, Lee LN, et al. Diagnosis of tuberculosis by an
enzyme-linked immunospot assay for interferon-c. Emerg Infect
Dis 2007; 13: 553–558.
44 Baba K, Sornes S, Hoosen AA, et al. Evaluation of immune responses
in HIV infected patients with pleural tuberculosis by the
56 Dogra S, Narang P, Mendiratta DK, et al. Comparison of a whole
blood interferon-c assay with tuberculin skin testing for the
detection of tuberculosis infection in hospitalized children in ruralIndia. J Infect 2007; 54: 267–276.
57 Dominguez J, Ruiz-Manzano J, De Souza-Galvao M, et al.
Comparison of two commercially available gamma interferon
blood tests for immunodiagnosis of tuberculosis. Clin Vaccine
Immunol 2008; 15: 168–171.
58 Goletti D, Carrara S, Vincenti D, et al. Accuracy of an immune
diagnostic assay based on RD1 selected epitopes for active
tuberculosis in a clinical setting: a pilot study. Clin Microbiol
Infect 2006; 12: 544–550.
59 Hansted E, Andriuskeviciene A, Sakalauskas R, et al. T-cell-baseddiagnosis of tuberculosis infection in children in Lithuania: a
country of high incidence despite a high coverage with bacille
Calmette-Guerin vaccination. BMC Pulm Med 2009; 9: 41.
60 Herrmann JL, Simonney N, Bergeron A, et al. IFNc and antibody
responses among French nurses during a tuberculosis contacttracing investigation. Pathol Biol (Paris) 2009; 57: e49–e53.
TUBERCULOSIS R. DIEL ET AL.
98 VOLUME 37 NUMBER 1 EUROPEAN RESPIRATORY JOURNAL
61 Hesseling AC, Mandalakas AM, Kirchner HL, et al. Highlydiscordant T cell responses in individuals with recent exposureto household tuberculosis. Thorax 2009; 64: 840–846.
62 Janssens JP, Roux-Lombard P, Perneger T, et al. Contribution of aIFN-c assay in contact tracing for tuberculosis in a low-incidence,high immigration area. Swiss Med Wkly 2008; 138: 585–593.
63 Kariminia A, Sharifnia Z, Aghakhani A, et al. Comparison ofQuantiFERON TB-G-test to TST for detecting latent tuberculosisinfection in a high-incidence area containing BCG-vaccinatedpopulation. J Eval Clin Pract 2009; 15: 148–151.
64 Laffitte E, Janssens JP, Roux-Lombard P, et al. Tuberculosisscreening in patients with psoriasis before antitumour necrosisfactor therapy: comparison of an interferon-c release assay vs.
tuberculin skin test. Brit J Dermatol 2009; 161: 797–800.65 Lien LT, Hang NT, Kobayashi N, et al. Prevalence and risk factors
for tuberculosis infection among hospital workers in Hanoi, VietNam. PLoS One 2009; 4: e6798.
66 Manuel O, Humar A, Preiksaitis J, et al. Comparison ofquantiferon-TB gold with tuberculin skin test for detecting latenttuberculosis infection prior to liver transplantation. Am JTransplant 2007; 7: 2797–2801.
67 Matulis G, Juni P, Villiger PM, et al. Detection of latent tuberculosisin immunosuppressed patients with autoimmune diseases: per-formance of a Mycobacterium tuberculosis antigen-specific inter-feron gamma assay. Ann Rheum Dis 2008; 67: 84–90.
68 Ozekinci T, Ozbek E, Celik Y. Comparison of tuberculin skin testand a specific T-cell-based test, T-Spot.TB, for the diagnosis oflatent tuberculosis infection. J Int Med Res 2007; 35: 696–703.
69 Perry S, Sanchez L, Yang S, et al. Reproducibility of QuantiFERON-TB gold in-tube assay. Clin Vaccine Immunol 2008; 15: 425–432.
70 Porsa E, Cheng L, Graviss EA. Comparison of an ESAT-6/CFP-10peptide-based enzyme-linked immunospot assay to a tuberculinskin test for screening of a population at moderate risk ofcontracting tuberculosis. Clin Vaccine Immunol 2007; 14: 714–719.
71 Ringshausen FC, Schlosser S, Nienhaus A, et al. In-hospital contactinvestigation among health care workers after exposure to smear-negative tuberculosis. J Occup Med Toxicol 2009; 4: 11.
72 Seyhan EC, Sokucu S, Altin S, et al. Comparison of theQuantiFERON-TB Gold In-Tube test with the tuberculin skin test
for detecting latent tuberculosis infection in hemodialysis patients.
Transpl Infect Dis 2010; 12: 98–105.
73 Soysal A, Turel O, Toprak D, et al. Comparison of positive
tuberculin skin test with an interferon-c-based assay in unexposed
children. Jpn J Infect Dis 2008; 61: 192–195.
74 Triverio PA, Bridevaux PO, Roux-Lombard P, et al. Interferon-c
release assays versus tuberculin skin testing for detection of latent
tuberculosis in chronic haemodialysis patients. Nephrol Dial
Transplant 2009; 24: 1952–1956.
75 Tsiouris SJ, Coetzee D, Toro PL, et al. Sensitivity analysis and
potential uses of a novel gamma interferon release assay for
diagnosis of tuberculosis. J Clin Microbiol 2006; 44: 2844–2850.
76 Vassilopoulos D, Stamoulis N, Hadziyannis E, et al. Usefulness of
enzyme-linked immunospot assay (Elispot) compared to tubercu-
lin skin testing for latent tuberculosis screening in rheumatic
patients scheduled for anti-tumor necrosis factor treatment.
J Rheumatol 2008; 35: 1271–1276.
77 Vinton P, Mihrshahi S, Johnson P, et al. Comparison of
QuantiFERON-TB Gold In-Tube Test and tuberculin skin test for
identification of latent Mycobacterium tuberculosis infection in
healthcare staff and association between positive test results and
known risk factors for infection. Infect Control Hosp Epidemiol 2009;
30: 215–221.
78 Goletti D, Butera O, Vanini V, et al. Response to Rv2628 latency
antigen associates with cured tuberculosis and remote infection.
Eur Respir J 2010; 36: 135–142.
79 Ferrara G, Losi M, D’Amico R, et al. Interferon-c-release assays
detect recent tuberculosis re-infection in elderly contacts. Int J
Immunopathol Pharmacol 2009; 22: 669–677.
80 Chee CB, KhinMar KW, Gan SH, et al. Latent tuberculosis infection
treatment and T-cell responses to Mycobacterium tuberculosis-
specific antigens. Am J Respir Crit Care Med 2007; 175: 282–287.
81 van Zyl-Smit RN, Pai M, Peprah K, et al. Within-subject variability
and boosting of T-cell interferon-c responses after tuberculin skin
testing. Am J Respir Crit Care Med 2009; 180: 49–58.
82 Krummel B, Strassburg A, Ernst M, et al. Potential role for IL-2
ELISpot in differentiating recent and remote infection in tubercu-
losis contact tracing. PloS ONE 2010; 5: e11670.
R. DIEL ET AL. TUBERCULOSIS
EUROPEAN RESPIRATORY JOURNAL VOLUME 37 NUMBER 1 99