A6 MMWR Update on IGRA in Diagnosis of TB 2010

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

1/28

department of health and human services

Centers for Disease Control and Prevention

Recommendations and Reports June 25, 2010 / Vol. 59 / No. RR-5

Morbidity and Mortality Weekly Reportwww.cdc.gov/mmwr

Updaed Guidelies r UsigIerer Gamma Release Assays

Deec Mycobacterium tuberculosisIeci Uied Saes, 2010

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

2/28

MMWR

Te MMWR series o publications is published by the Oce oSurveillance, Epidemiology, and Laboratory Services, Centers orDisease Control and Prevention (CDC), U.S. Department o Healthand Human Services, Atlanta, GA 30333.

Suggested Citation: Centers or Disease Control and Prevention.[itle]. MMWR 2010;59(No. RR-#):[inclusive page numbers].

Ceers r Disease Crl ad PreveiTomas R. Frieden, MD, MPH

Director

Harold W. Jae, MD, MAAssociate Director or Science

James W. Stephens, PhDOce o the Associate Director or Science

Stephen B. Tacker, MD, MScDeputy Director or

Surveillance, Epidemiology, and Laboratory Services

Edirial ad Prduci SaFrederic E. Shaw, MD, JD

Editor, MMWRSeriesChristine G. Casey, MD

Deputy Editor, MMWRSeries

eresa F. RutledgeManaging Editor, MMWRSeries

David C. JohnsonLead echnical Writer-Editor

Jerey D. Sokolow, MAProject Editor

Martha F. BoydLead Visual Inormation Specialist

Malbea A. LaPeteStephen R. Spriggserraye M. Starr

Visual Inormation Specialists

Quang M. Doan, MBAPhyllis H. King

Inormation echnology Specialists

Edirial BardWilliam L. Roper, MD, MPH, Chapel Hill, NC, Chairman

Virginia A. Caine, MD, Indianapolis, INJonathan E. Fielding, MD, MPH, MBA, Los Angeles, CA

David W. Fleming, MD, Seattle, WAWilliam E. Halperin, MD, DrPH, MPH, Newark, NJ

King K. Holmes, MD, PhD, Seattle, WADeborah Holtzman, PhD, Atlanta, GA

John K. Iglehart, Bethesda, MD

Dennis G. Maki, MD, Madison, WIPatricia Quinlisk, MD, MPH, Des Moines, IAPatrick L. Remington, MD, MPH, Madison, WI

Barbara K. Rimer, DrPH, Chapel Hill, NCJohn V. Rullan, MD, MPH, San Juan, PR

William Schaner, MD, Nashville, NAnne Schuchat, MD, Atlanta, GA

Dixie E. Snider, MD, MPH, Atlanta, GAJohn W. Ward, MD, Atlanta, GA

ContEntS

Introduction .............................................................................. 1

Methods or Updating IGRA Guidelines ...................................... 2

Background .............................................................................. 2

The Epidemiology o Tuberculosis and M. tuberculosis Inection . 2

Development o Intereron Gamma Release Assays (IGRAs)and Interpretation Criteria .................................................... 2

FDA-Approved Intended Use or IGRAs ................................... 5

Assessment o QFT-GIT and T-Spot Accuracy, Specifcity,and Sensitivity ..................................................................... 5

Use o QFT-GIT and T-Spot in Contact Investigations ................. 7

Value o QFT-GIT and T-Spot in Predicting Subsequent Active Tuberculosis .............................................................. 7

Use o QFT-GIT and T-Spot or Testing Children ........................ 8

Use o QFT-GIT and T-Spot or Testing ImmunocompromisedPersons ............................................................................... 9

Considerations or Programs ................................................... 9

Recommendations ................................................................... 10

General Recommendations or Use o IGRAs .......................... 10

Test Selection ........................................................................ 10

Medical Management Ater Testing ........................................ 12

Areas or Additional Research .................................................. 12

Reerences .............................................................................. 13

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

3/28

Vol. 59 / RR-5 Recommendations and Reports 1

Updaed Guidelies r Usig Ierer Gamma ReleaseAssays Deec Mycobacterium tuberculosisIeci

Uied Saes, 2010Prepared by

Gerald H. Mazurek, MD, John Jereb, MD, Andrew Vernon, MD, Phillip LoBue, MD, Stean Goldberg, MD, Kenneth Castro, MDDivision o uberculosis Elimination, National Center or HIV, SD, and B Prevention, CDC

Summary

In 2005, CDC published guidelines or using the QuantiFERON-B Gold test (QF-G) (Cellestis Limited, Carnegie, Victoria,Australia) (CDC. Guidelines or using the QuantiFERON-B Gold test or detectingMycobacterium tuberculosis inection,United States. MMWR;54[No. RR-15]:4955). Subsequently, two new intereron gamma (IFN-) release assays (IGRAs) wereapproved by the Food and Drug Administration (FDA) as aids in diagnosingM. tuberculosis inection, both latent inection andinection maniesting as active tuberculosis. Tese tests are the QuantiFERON-B Gold In-ube test (QF-GI) (Cellestis Limited,Carnegie, Victoria, Australia) and the -SPO.B test (-Spot) (Oxord Immunotec Limited, Abingdon, United Kingdom). Teantigens, methods, and interpretation criteria or these assays dier rom those or IGRAs approved previously by FDA.

For assistance in developing recommendations related to IGRA use, CDC convened a group o experts to review the scientifc

evidenceand provide opinions regarding use o IGRAs. Data submitted to FDA, published reports, and expert opinion relatedto IGRAs were used in preparing these guidelines. Results o studies examining sensitivity, specifcity, and agreement or IGRAsand S vary with respect to which test is better. Although data on the accuracy o IGRAs and their ability to predict subsequentactive tuberculosis are limited, to date, no major defciencies have been reported in studies involving various populations.

Tis report provides guidance to U.S. public health ocials, health-care providers, and laboratory workers or use o FDA-approved IGRAs in the diagnosis oM. tuberculosis inection in adults and children. In brie, Ss and IGRAs (QF-G, QF-GI, and -Spot) may be used as aids in diagnosingM. tuberculosis inection. Tey may be used or surveillance purposes and toidentiy persons likely to beneft rom treatment. Multiple additional recommendations are provided that address quality control,test selection, and medical management ater testing.

Although substantial progress has been made in documenting the utility o IGRAs, additional research is needed that ocuseson the value and limitations o IGRAs in situations o importance to medical care or tuberculosis control. Specifc areas needingadditional research are listed.

Te material in this report originated in the National Center or HIV,SD, and B Prevention, Kevin Fenton, MD, PhD, Director; andthe Division o uberculosis Elimination, Kenneth G. Castro, MD,Director.Corresponding preparer: Gerald H. Mazurek, MD, Division ouberculosis Elimination, National Center or HIV, SD, and BPrevention, CDC, 1600 Cliton Rd., N.E., MS E-10, Atlanta, GA30333. elephone: 404-639-8174; Fax: 404-639-8961; E-mail:[email protected].

IrduciBeore 2001, the tuberculin skin test (S) was the only

practical and commercially available immunologic test orMycobacterium tuberculosis inection approved in the UnitedStates (1). Recognition that intereron gamma (IFN-) playsa critical role in regulating cell-mediated immune responsestoM. tuberculosisinection led to development o intererongamma release assays (IGRAs) or the detection oM. tuberculo-sisinection (24). IGRAs detect sensitization toM. tuberculosis

by measuring IFN- release in response to antigens representingM. tuberculosis. In 2001, the QuantiFERON-B test (QF)(Cellestis Limited, Carnegie, Victoria, Australia) became therst IGRA approved by the Food and Drug Administration(FDA) as an aid or diagnosingM. tuberculosisinection (5,6)In 2005, the QuantiFERON-B Gold test (QF-G) (CellestisLimited, Carnegie, Victoria, Australia) became the secondIGRA approved by FDA as an aid or diagnosingM. tuberculosisinection (7,8). CDC published guidelines or using QF in2003 and or using QF-G in 2005 (6,8).

Updated IGRA guidelines are needed because since 2005two new IGRAs have been approved by FDA, and severalhundred peer-reviewed articles describing clinical studies oIGRAs have been published. Tis report provides updatedguidance to U.S. public health ocials, health-care providersand laboratory workers or use o FDA-approved IGRAs in thediagnosis oM. tuberculosisinection in adults and children.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

4/28

2 MMWR June 25, 2010

Mehds r Updaig IGRAGuidelies

CDC identied relevant reports published through August2008 by searching PubMed or articles written in English thatlisted tuberculosis as the major MeSH topic and that included

either QuantiFERON or -Spot in the title or abstract.CDC identied additional published reports by contactingtest manuacturers and examining reerences listed in retrievedarticles. Tese search methods identied 152 potentially rel-evant articles. CDC reviewed the methods used in each studyto select 96 primary reports that provided data related to 1)sensitivity or specicity o QF-GI or -Spot; 2) agreemento QF-GI and -Spot results with each other or with Sresults; 3) association o QF-GI or -Spot results with riskorM. tuberculosisinection or subsequent active tuberculosis;or 4) evaluation o QF-GI or -Spot use in contact investi-gations, immunocompromised persons, or children.

During August 45, 2008, CDC convened a meetingin Atlanta, Georgia, to consider the use o QF-GI and-Spot in U.S. tuberculosis-control activities. At this meeting,tabulated study results, descriptive summaries, explanationsby study authors, and commentaries rom test manuacturerswere presented to an Expert Committee* comprising tuber-culosis-control ocials, clinicians, laboratorians, and leadingresearchers with IGRA expertise, together with representa-tives o the American Academy o Pediatrics, the AmericanToracic Society, the Advisory Council or the Elimination ouberculosis, the Association o Public Health Laboratories,CDC, FDA, the Inectious Disease Society o America, theNational uberculosis Controllers Association, Stop B USA,the U.S. Army, the U.S. Air Force, and the Veterans HealthAdministration. Data rom most o the 96 primary reports usedby CDC as the evidence on which these guidelines are basedwere available or review by the expert committee either as pub-lished articles or articles accepted or publication. CDC askedmembers o the Expert Committee to provide written opinionsregarding how FDA-approved IGRAs should be used.

CDC used the published reports, data submitted to FDA, theproduct package inserts, and expert opinion related to QF-GI and -Spot to prepare these guidelines. CDC coordinated

development o these guidelines with the American Academyo Pediatrics, the American Toracic Society, and the InectiousDisease Society o America.

Backgrud

the Epidemilgy tuberculsis adM. tuberculosisIeci

Globally, nine million persons develop active disease attrib-

utable to M. tuberculosisinection annually, and one thirdo the worlds population, approximately 2 billion personsare thought to be latently inected with M. tuberculosis(9)Although persons with latentM. tuberculosisinection (LBI)do not maniest overt symptoms o active tuberculosis andare not inectious, they are at increased risk or developingactive disease and becoming inectious. Approximately twomillion persons die each year rom active tuberculosis despitethe existence o eective treatments or both latent inectionand active disease.

Te prevalence o active tuberculosis in the United States hasdeclined rom 6.2 cases per 100,000 persons in 1998 to 4.2

cases per 100,000 persons in 2008 (10). During 19982007o the 153,555 persons in the United States who had received adiagnosis o active tuberculosis, 3,708 (2.4%) died beore treat-ment or active tuberculosis was started, and 10,777 (7.0%)died ater starting treatment but beore treatment was com-pleted (CDC, unpublished data, 2008). A S survey in 2000indicated that an estimated 11,213,000 U.S. residents (4.2%o the civilian, noninstitutionalized U.S. population aged >1year) had LBI, representing a 60% decline rom 1972 (11)However, the declines were not uniorm among all segmentso the U.S. population, and rates oM. tuberculosisinectionand active tuberculosis vary considerably. Categorization o

the risk or inection (Box 1) and or progression to activedisease (Box 2) acilitates targeted testing and selection o thosepersons likely to benet rom treatment or latent inection(12). Identication o persons who are at increased risk or apoor clinical outcome (e.g., meningitis, disseminated diseaseor death) i active tuberculosis occurs (Box 2) is an importantcomponent o targeted testing and treatment. U.S. residentswith none o the recognized risk characteristics are consideredto be at low risk or both inection and disease romM. tuber-culosis. Te prevalence oM. tuberculosisinection among suchpersons is estimated to be 1% (11).

Develpme Ierer GammaRelease Assays (IGRAs) adIerpreai Crieria

Ss have been used worldwide or more than a century as anaid in diagnosing both LBI and active tuberculosis. A positiveS result is associated with an increased risk or current oruture active tuberculosis (1316). However, certain limitations

* he names o the members o the IGRA Expert Committee and the IGRAExpert Committee presenters appear on page 25 o this report.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

5/28

Vol. 59 / RR-5 Recommendations and Reports 3

are associated with the use o Ss. A valid S requires properadministration by the Mantoux method with intradermal injec-tion o 0.1mL o tuberculin-puried protein derivative (PPD)into the volar surace o the orearm. In addition, patients mustreturn to a health-care provider or test reading, and inaccura-cies and bias exist in reading the test. Also, alse-positive Sscan result rom contact with nontuberculous mycobacteria orvaccination with Bacille Calmette-Guerin (BCG), because theS test material (PPD) contains antigens that are also in BCGand certain nontuberculous mycobacteria (13,17,18).

In 2001, QF became the rst IGRA approved by FDAas an aid or diagnosingM. tuberculosisinection (5,6). Tistest used an enzyme-linked immunosorbent assay (ELISA) tomeasure the amount o IFN- released in response to PPDcompared with controls. CDC issued guidelines on the useo QF in 2003 (6). However, QF specicity was less thanthat o S despite the use oM. avium antigen as a controlor nontuberculous mycobacterial sensitization and saline as

a negative control (19). QF has not been available commer-cially since 2005.

o improve specicity, new IGRAs were developed. Tese

IGRAs assess response to synthetic overlapping peptides thatrepresent specicM. tuberculosisproteins, such as early secre-tory antigenic target-6 (ESA-6) and culture ltrate protein 10(CFP-10). Tese proteins are present in allM. tuberculosisandthey stimulate measurable release o IFN- in most inectedpersons, but they are absent rom BCG vaccine strains androm most nontuberculous mycobacteria (20). Tus, as tesantigens, these proteins oer improved test specicity com-

BOX 1. Risk actors or Mycobacterium tuberculosisinection

Persons at increased risk* orM. tuberculosisinectionclose contacts o persons known or suspected tohave active tuberculosis;

oreign-born persons rom areas that have a highincidence o active tuberculosis (e.g., Arica, Asia,Eastern Europe, Latin America, and Russia);persons who visit areas with a high prevalence oactive tuberculosis, especially i visits are requentor prolonged;residents and employees o congregate settingswhose clients are at increased risk or active tuber-culosis (e.g., correctional acilities, long-term careacilities, and homeless shelters);health-care workers who serve clients who are atincreased risk or active tuberculosis;populations dened locally as having an increasedincidence o latent M. tuberculosisinection oractive tuberculosis, possibly including medicallyunderserved, low-income populations, or personswho abuse drugs or alcohol; andinants, children, and adolescents exposedto adults who are at increased risk or latentM. tuberculosisinection or active tuberculosis.

Source: Based on CDC. argeted tuberculin testing and treatment olatent tuberculosis inection. MMWR 2000;49(No. RR-6).* Persons with these characteristics have an increased risk orM. tuberculosis

inection compared with persons without these characteristics.

BOX 2. Risk actors or progression o inection to activetuberculosis

Persons at increased risk* or progression o inection toactive tuberculosis include

persons with human immunodeiciency virus

(HIV) inection;inants and children aged

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

6/28

4 MMWR June 25, 2010

pared with PPD. However, ESA-6 and CFP-10 are presentinM. kansasii, M. szulgai, andM. marinum , and sensitizationto these organisms might contribute to the release o IFN-in response to these antigens and cause alse-positive IGRAresults. Because ESA-6 and CFP-10 are recognized by ewer lymphocytes and stimulate less IFN- release compared

with PPD, a more sensitive ELISA than was used or QF isrequired to measure IFN- concentrations and responses toESA 6 and CFP-10.

In 2005, the QuantiFERON-B Gold test (QF-G) (CellestisLimited, Carnegie, Victoria, Australia) became the second IGRAapproved by FDA as an aid or diagnosingM. tuberculosisinec-tion (7,8). Itassesses the immunologic responsiveness o testedpatients to ESA-6 and CFP-10. For QF-G, separate aliquotso resh whole blood are incubated with controls and with twoseparate mixtures o peptides, one representing ESA-6 and theother representing CFP-10. Te amount o IFN- released inresponse to ESA-6 or CFP-10 (i.e., the ESA-6 Response orthe CFP-10 Response) is calculated as the dierence in IFN-concentration in plasma rom blood stimulated with antigenminus the IFN- concentration in plasma rom blood incubatedwith saline (i.e., Nil). For QF-G, the B Response is the highero the ESA-6 Response or the CFP-10 Response. A stipulationor FDA approval was inclusion o interpretation criteria thataddressed the potential or alse-positive results accompanyinghigh Nil values (i.e., >0.7 IU/ml).

In 2005, CDC issued guidelines or using QF-G (8), butthe criteria that addressed interpretation when Nil valuesare high were subsequently revised (able 1) (21). Te 2005

QF-G guidelines indicated that QF-G may be used in allcircumstances in which a S was recommended, includingcontact investigations, evaluation o recent immigrants, andserial-testing surveillance programs or inection control (e.g.,those or health-care workers) (8). Te guidelines providedcautions or testing persons rom selected populations, includ-ing persons at increased risk or progression to active diseasei inected.

For IGRAs to measure IFN- response accurately, a reshblood specimen that contains viable white blood cells is needed.Tis requirement limited the use o early IGRAs to acilities inwhich trained laboratorians could begin testing blood within

a ew hours o its collection. Te QuantiFERON-B GoldIn-ube test (QF-GI) (Cellestis Limited, Carnegie, Victoria,Australia) was developed to address this limitation. In October2007, QF-GI became the third IGRA approved by FDA asan aid or diagnosingM. tuberculosisinection (22). Controlmaterials and antigens or QF-GI are contained in specialtubes used to collect blood or the test, thus allowing moredirect testing o resh blood. One tube contains test antigensthat consist o a single mixture o 14 peptides representing

the entire amino acid sequences o ESA-6 and CFP-10 andpart o the sequence o B7.7. Te two accompanying tubesserve as negative and positive controls: the negative-controtube contains heparin alone, and the positive-control tubecontains heparin, dextrose, and phytohemaglutinin. Blood (1ml) is collected into each o the three tubes, mixed with the

reagents already in the tubes, and incubated or 1624 hoursPlasma is separated, and the IFN- concentration in the plasmais determined using the same sensitive ELISA used or QF-Go interpret QF-GI as approved by the FDA (able 2), theB Response is calculated as the dierence in IFN- concentra-tion in plasma rom blood stimulated with antigen (i.e., thesingle cocktail o peptides representing ESA-6, CFP-10, andB7.7) minus the IFN- concentration in plasma rom bloodincubated without antigen (i.e., Nil).

QF-GI was evaluated in the United States and used inother countries prior to FDA approval in 2007, and users othe test promulgated a variety o interpretation criteria. Somepublished reports used criteria or QF-GI that were similarto those being used or QF-G. As compared with FDA-approved QF-G interpretation criteria (able 1), the FDAcriteria approved or QF-GI in 2007 (able 2) interprettests with a Nil o 0.78.0 and a B Response o 25%50%o Nil as positive rather than as indeterminate. Also, tests witha Nil o 0.78.0 and a B Response that is

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

7/28

Vol. 59 / RR-5 Recommendations and Reports 5

FDA-Apprved Ieded User IGRAs

FDA has approved both QF-GI and -Spot as in vitrodiagnostic aids or detection o M. tuberculosisinection(22,23). Both tests are approved as indirect tests orM. tuber-

culosisinection (including inection resulting in active disease)when used in conjunction with risk assessment, radiography,and other medical and diagnostic evaluations. Te FDA-approved indications or QF-GI and -Spot are similar toindications or QF-G and S using either ubersol PPD(Sano Pasteur Ltd., oronto, Ontario, Canada) or AplisolPPD (JHP Pharmaceuticals, LLC, Rochester, Michigan).Because QF-G, QF-GI, -Spot, and S each measuredierent aspects o the immune response and use dierentantigens and interpretation criteria, test results might not beinterchangeable. Dierent tests can yield dierent results.

Assessme QFt-GIt ad t-SpAccuracy, Specifciy, ad Sesiiviy

Limiais i Assessig Accuracy

Assessments o accuracy o tests orM. tuberculosisinectionare hampered by the lack o conrmatory tests to diagnoseLBI and culture-negative active tuberculosis. Accuracy is ameasure o the proportion o test results that are correct andencompasses assessment o specicity (the proportion o truenegatives that have negative test results) and sensitivity (theproportion o true positives that have positive test results).Assessments o accuracy o tests or

M. tuberculosisinection are

dicult because there is no gold standard to conrm a diag-nosis o LBI or culture-negative active tuberculosis. However,approximations o accuracy, sensitivity, and specicity can bemade by testing populations with known characteristics. Forexample, to assess the sensitivity o IGRAs, researchers canobserve the proportion o positive IGRA results among personswith culture-conrmed active tuberculosis, a group or whomthe IGRA should be positive (i.e., true positives). Likewise,to assess the specicity o IGRAs, researchers can observe theproportion o negative IGRA tests among persons who arevery unlikely to have M. tuberculosisinection (i.e., assumed

negatives). Researchers also can characterize actors associatedwith discordance between dierent tests or conduct ollow-upstudies to determine the subsequent rate o active tuberculosisor persons with positive or negative IGRA results. However,although sensitivity and specicity are inherent character-istics o the tests, with no gold standard, estimates o testperormance might fuctuate as a result o dierences in thestudy population and the rate o diagnostic misclassication(e.g., as a result o dierences in prevalence M. tuberculosis

and nontuberculous mycobacterial inection, malnutritionand immune suppression). In addition, because Ss andIGRAs are indirect tests that measure immunologic responsesand are not direct tests that detect the causative organism orcomponents o the organism, assessments o sensitivity amongpersons with culture-conrmed active tuberculosis might not

provide reliable estimates o sensitivity or LBI. Immunologicdierences that allow progression o inection to disease mighaect immunologic test results. In addition, treatment can alterimmunologic responses and might alter test results. Estimateso specicity among low-risk populations might underestimatespecicity because some persons might have inection resultingrom unrecognized exposure.

Assessment o test accuracy is complicated urther by the use odierent test methods and interpretation criteria or S, QF-GI, and -Spot in published reports. Most published reportsevaluating QF-GI or -Spot accuracy (2653) (ables 47)have used interpretation criteria dierent rom those approvedby FDA. Also, in published studies in which IGRA results havebeen compared with S results (27,28,3141,4345,49,50)the S antigens and cut points in indurations used to separatenegative and positive results diered. In addition, or evaluationo QF-GI, some investigators used methods that did notinclude a positive control or QF-GI (28,30), in contrast tothe methods approved by FDA. Inclusion o a positive controlincreases estimates o sensitivity by excluding indeterminateresults with low Mitogen Responses, which otherwise mightbe interpreted as negative. For example, i blood samples areprocessed improperly to the point that they lose the ability to

produce IFN- and a positive control is not used, the IGRAresults or these samples will be interpreted as negative. With apositive control, they will be interpreted as indeterminate andnot be included in the calculations o sensitivity (i.e., they willbe removed rom the denominator). Tis is similar to excludingpersons who do not return to have their S read rom estimateo S sensitivity.

Incorporation o a borderline category or the -Spot asapproved by FDA (able 3) increases test accuracy by clas-siying results near the cut point (at which small variationsmight aect the interpretation) as neither positive or negativeAlthough not included in FDA-approved interpretation criteria

or QF-GI (able 2), an appropriate borderline categoryor QF-GI might increase its accuracy or the same reasonsAnother tactic or improving detection sensitivity is to useany positive result rom multiple tests, as is done with cultureor nucleic acid amplication tests. Interpreting any positiveresult rom multiple tests as evidence o inection typicallyincreases detection sensitivity and decreases specicity. On theother hand, requiring positive results rom two or more tests

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

8/28

6 MMWR June 25, 2010

typically has the opposite eect (i.e., decreasing sensitivity andincreasing specicity).

Esimaes Sesiiviy

Estimates o QF-GI and -Spot sensitivity have varied widely in published studies (ables 4 and 5), which have

involved predominantly adults with culture-conrmed activetuberculosis. In general, QF-GI and -Spot sensitivitiesare considered similar to those or S. However, caution isrequired when comparing test sensitivity rom these studiesbecause 1) some cohorts were not limited to subjects withmicrobiologically conrmed active tuberculosis (and in real-ity might not have had active tuberculosis); 2) in the majorityo studies, head-to-head comparisons o IGRAs were notperormed in the same subjects; and 3) test methods andinterpretation criteria used in reported studies oten dieredrom those approved by FDA.

When data rom published studies related to QF-GI

sensitivity in patients with culture-conrmed active tuberculosis(2630,32,33,35,3739) were pooled (able 4) and sensitivitywas determined as the number o subjects with positive QF-GI results divided by the number with positive or negativeresults, pooled QF-GI sensitivity was 81%, compared with70% reported by a study that estimated sensitivity on the basiso a meta-analysis (54). In studies that compared the sensitivityo QF-GI to that o S in patients with culture-conrmedactive tuberculosis (27,28,32,33,35,3739), pooled QF-GIsensitivity was 83% and pooled S sensitivity was 89%. Inthe 11 studies that compared QF-GI and S in patients inwhom active tuberculosis (not necessarily culture-conrmed) wasdiagnosed, six studies (28,32,34,3739) demonstrated no statis-tically signicant dierence between the two tests, three studies(27,31,33) demonstrated greater sensitivity or S, and twostudies (35,36) demonstrated greater sensitivity or QF-GI.

When data rom published studies related to -Spot sensi-tivity in patients with culture-conrmed active tuberculosis(28,33,38,42,46,48,5052) were pooled (able 5), and sensi-tivity was determined as the number o subjects with positive-Spot results divided by the number with positive or negativeresults, pooled -Spot sensitivity was 91%. In studies thatcompared the sensitivity o -Spot to that o S in patients

with culture-conrmed tuberculosis (28,33,38,39,50), pooled-Spot sensitivity was 90% and S sensitivity was 89%. Inthe 12 studies that compared -Spot and S sensitivity inpatients diagnosed with active tuberculosis (not necessarily cul-ture-conrmed), nine demonstrated no statistically signicantdierence in the two tests (28,31,33,38,39,43,44,49,50), andthree demonstrated greater sensitivity or -Spot (39,40,44).

In three published studies that evaluated S, QF-GI, and-Spot (28,33,39), pooled sensitivity or S, -Spot, and QF-

GI were 95%, 91%, and 84%, respectively. Te largest o thesestudies was conducted in Singapore and involved more than 270persons with culture-conrmed active tuberculosis (33). In thastudy, the estimates o sensitivity o -Spot and o S (using a10-mm cuto) were similar (94% and 95% respectively; p=0.84)and signicantly greater than QF-GI (83%; p

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

9/28

Vol. 59 / RR-5 Recommendations and Reports 7

inection and the proportion o inections that are conrmedmicrobiologically, estimates o recent and remote exposure, age,race, prior BCG vaccination, recent S, and coexisting diseases,including nontuberculous mycobacterial inection and condi-tions with immunosupression (e.g., human immunodeciencyvirus [HIV] inection). Increasing age is a risk orM. tuberculosis

inection because o longer time or potential exposure andbecause older persons might have been alive when tuberculosiswas more prevalent. Te association o older age with positiveS and IGRA results generally is attributed toM. tuberculosisinections that accumulate over time. Te observation in somestudies that increasing age is associated more strongly with Sresults than with IGRA results suggests that a S might bemore sensitive than IGRAs in detecting remote inections thatoccurred years earlier (58,61).

Investigations examining the eect o PPD injection on sub-sequent IGRAs have produced conficting results (59,6266);outcome dierences probably are attributable to dierencesin the study population (inected versus noninected subjects,recent versus temporally remote inection, and risk or ongoingexposure), timing o IGRA testing ater PPD injection, the IGRAormat, and the denition o boosting used. PPD injection shouldbe expected to boost anamnestic immune responses measured byIGRA originating romM. tuberculosisinection, but not romBCG vaccination or in nonsensitized persons. Additional studiesexamining the eect o PPD injection on IFN- responses areneeded to dene the requency, magnitude, induction time, andlongevity o IGRA boosting ollowing a S.

Uncertainty exists regarding the reproducibility o IGRA

results in individual patients and the clinical signicance ofuctuations in measured IFN- responses. Longitudinal studieshave revealed considerable fuctuation in IFN- responses withserial testing in individual patients (59,62,63,65,6771). Tesefuctuations might be attributed to limitations in the precisiono IGRAs or to actual fuctuations in IFN- responses in thepatient. Some increases in IFN- response might be attributedto new inection or boosting ollowing a S. Some decreasesin IFN- response in individual persons might be attributedto antimycobacterial treatment. However, or the most part,fuctuations in IFN- responses among serially tested individualpatients reported in longitudinal studies remain unexplained

and nonspecic. Te magnitude o these fuctuations can beo sucient size to cause test interpretations to change romnegative to positive (conversion) or rom positive to negative(reversion), especially when the IFN- responses are near cutpoints separating positive and negative results. Well-controlledstudies are needed to urther dene the causes o individualvariations in IFN- response and to develop criteria to di-erentiate nonspecic variation rom that associated with newor resolving inection.

Use QFt-GIt ad t-Sp i CacIvesigais

Several reports o contact investigations have included results romQF-GI and -Spot (able 8) (30,31,58,61,7274). In two othese investigations (58,73), greater recent exposure (as measured

by duration o exposure or inectiousness o the source based on ahigher number o acid-ast bacilli in their sputa) was more stronglyassociated with positive IGRA results than with positive S resultssuggesting that IGRAs might be better than the S at detectingrecent inection. In these studies, persons with lower amounts orecent exposure were more likely to be positive by S than IGRAsuggesting that the S might have been better than the IGRAs atdetecting remote inection that was present prior to (and thereoredid not occur as a result o) the recent exposure (58). In two otherinvestigations (72,74), neither S nor IGRA results were associ-ated with measures o recent exposure. In another investigation (30)the proximity o recent exposure (i.e., same room, dierent room,

or dierent house) was more strongly associated with S resultsthan QF-GI results.

Value QFt-GIt ad t-Sp iPredicig Subseque Acivetuberculsis

O critical importance, is a tests ability to predict risk orsubsequent active tuberculosis. For a person with a positiveS, the lietime risk or active tuberculosis is estimated to be5%10% (16,75). However, very ew longitudinal data existon the ability o IGRAs to predict risk or subsequent active

tuberculosis.In one study in Germany involving 601 close contacts o

persons with smear-positive, culture-conrmed active tuber-culosis, QF-GI was reported to perorm better than aS using a 5 mm cut point in predicting subsequent activetuberculosis (76). Whereas ve (2.3%) o 219 contacts withS induration 5 mm developed tuberculosis, six (14.6%)o 41 contacts with positive QF-GI results developed thedisease (p=0.003). However, an unusually large proportion(59%) o the contacts had S induration that ranged rom 5mm to 9 mm. Te proportion o those considered positive byS using a 10 mm cuto who developed active tuberculosis(ve o 90 [5.6%]) was similar to the proportion positive byQF-GI (six o 41 [14.6%]; p=0.1). In addition, only two othe six contacts with positive QF-GI results who developedactive tuberculosis had the diagnosis conrmed by culture. Asnoted in a published comment on the article, the sensitivityor predicting subsequent active tuberculosis did not diersignicantly or the two tests (77). Te QF-GI sensitivitywas 100% (95% condence interval [CI] = 54%100%) and

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

10/28

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

11/28

Vol. 59 / RR-5 Recommendations and Reports 9

concentration used or stimulation and dierences in interpre-tation criteria can aect the number o indeterminate results,especially when dierent IGRA ormats are compared. Tird,concerns relate to diculties in collecting blood or these testsand the need or a relatively large volume o blood rom smallchildren (especially or inants). Finally, certain pediatricians

have expressed concern that IGRAs might have lower sensitivitythan Ss in children (81,90,91).

In general, sensitivity o IGRAs in children is expected to becomparable to S. In one study o 28 children with culture-conrmed active tuberculosis who were aged 4 months7years, estimates o sensitivity or S, QF-GI, and -Spot were comparable at 100%, 93%, and 93% respectively(p=0.15) (28). Sensitivities o these tests were also similar inanother study o nine children who had active tuberculosis;six(67%) were positive by -Spot, six (67%) were positive byQF-GI, and nine (100%) were positive by S (31). Inanother study involving 25 children with culture-conrmedactive tuberculosis, estimates o sensitivity were 88% or Sat 10 mm and 83% or S at 15 mm, 80% or QF-GI,and 58% or -Spot (39). In the same study, when childrenwith probable active tuberculosis were included (dened onthe basis o epidemiologic, clinical, and radiographic ndingsin the absence o a positive culture), sensitivity or S at 10mm ell to 71%, sensitivity or S at 15 mm ell to 60%,and sensitivity or QF-GI and -Spot ell to 64% and50%, respectively. However, the methods used or diagnosingactive tuberculosis in this study were not stated specicallyand might have included use o S results. In another study

that evaluated 154 children aged 515 years with culture-conrmed active tuberculosis, results indicated that S wasmore sensitive than QF-GI (90% and 76%, respectively;p0.99) (27). QF-GI sensitivity was not

signicantly dierent among persons with HIV inection thanamong those without inection (81% and 73%, respectivelyp=0.59). In another study in Zambia involving 112 persons(59 were inected with HIV, 37 were not inected with HIV,and 16 were not tested) in whom active tuberculosis wasdiagnosed on the basis o sputum smear (36), QF-GI and

S were signicantly less sensitive in persons inected withHIV than in persons not inected with HIV (76% compared with 97% or QF-GI; p=0.02 and 55% compared with81% or S, p=0.04). Among persons with HIV inectionQF-GI sensitivity tended to be higher than S sensitivity(76% and 55%, respectively; p=0.06). However, in this studyreduced S sensitivity might have resulted rom delayedreading o Ss, which were read 48164 hours ater PPDinjection. Low CD4 counts were associated with increases inalse-negative S results and indeterminate and alse-negativeQF-GI results.

Published comparisons have not demonstrated signicantdierences in the proportion o positive QF-GI results ascompared with the proportion o positive S results amongHIV-inected persons screened or M. tuberculosisinection(9396). QF-GI results rom two studies suggest that theproportion o indeterminate QF-GI results among HIV-inected persons (17% and 19%, respectively) is similar to theproportion among uninected persons (14% and 0, respec-tively; p=0.88 and p=0.18, respectively) (27,36). Howeverin another study among HIV-inected persons, CD4 countswere lower in those with indeterminate QF-GI results ascompared with those with positive or negative results (p

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

12/28

10 MMWR June 25, 2010

ingM. tuberculosisinection. Unlike Ss, IGRA results can beavailable within 24 hours without the need or a second visit. Aslaboratory-based assays, IGRAs are not subject to the biases anderrors associated with S placement and reading. However,errors in collecting, labeling, or transporting blood specimens,or while perorming and interpreting these assays can decrease

IGRA accuracy. Also, availability o IGRAs is limited by the needor a resh blood sample and the potential or delays as a resulto the long distances to laboratories that oer these tests.

Te cost or an IGRA is substantially greater than that or aS (109). However, this additional cost might be oset bydecreases in the number o persons testing positive and theassociated costs o evaluating and treating persons with positivetest results (110). Use o an IGRA might increase acceptance otreatment or LBI (111). However, cost-eectiveness studiesare limited by the lack o critical data on the relative ability othese tests to predict subsequent disease.

Recmmedais

Geeral Recmmedais rUse IGRAs

Ss and IGRAs (QF-G, QF-GI, and -Spot) shouldbe used as aids in diagnosing inection withM. tuberculosis.Tese tests may be used or surveillance purposes or toidentiy persons likely to benet rom treatment, includ-ing persons who are or will be at increased risk or M.tuberculosisinection (Box 1) or or progression to active

tuberculosis i inected (Box 2).IGRAs should be perormed and interpreted according toestablished protocols using FDA-approved test ormats.Tey should be perormed in compliance with ClinicalLaboratory Improvement Amendment (CLIA) standards.Both the standard qualitative test interpretation andthe quantitative assay measurements should be reportedtogether with the criteria used or test interpretation. Tiswill permit more rened assessment o results and promoteunderstanding o the tests.Arrangement or IGRA testing should be made priorto blood collection to ensure that the blood specimen

is collected in the proper tubes, and that testing can beperormed within the required timerame.Prior to implementing IGRAs, each institution and tuber-culosis-control program should evaluate the availability,overall cost, and benets o IGRAs or their own setting.In addition, programs should consider the characteristicso the population to be tested.As with the S, IGRAs generally should not be used ortesting persons who have a low risk or both inection and

progression to active tuberculosis i inected (except or thoselikely to be at increased risk in the uture). Screening suchpersons diverts resources rom higher priority activities andincreases the number o alse-positive results. Even with atest specicity approaching 99%, when the prevalence oM. tuberculosisinection is 1%, the majority o positive

results will be alse positives. I persons at low risk or bothinection and progression are to be tested, selection o thetest with the greatest specicity will minimize alse-positiveresults, reduce unnecessary evaluation and treatment, andminimize the potential or adverse events rom unnecessarytreatment.

tes SeleciSelection o the most suitable test or combination o testsor detection oM. tuberculosisinection should be madeon the basis o the reasons and the context or testing

test availability, and overall cost eectiveness o testingResults o studies examining sensitivity, specicity, andagreement or IGRAs and S vary with respect to whichtest is better. Although data on the accuracy o IGRAs andtheir ability to predict subsequent active tuberculosis arelimited, to date, no major deciencies have been reportedin studies involving various populations. As use o thesetests increases, greater understanding o their value andlimitations will be gained.An IGRA may be used in place o (but not in addition to) aS in all situations in which CDC recommends tuberculinskin testing as an aid in diagnosingM. tuberculosisinection

with preerences and special considerations noted belowDespite the indication o a preerence in these instancesuse o the alternative test (FDA-approved IGRA or S)is acceptable medical and public health practice.

Siuais i Which a IGRA Is Preerred Bua tSt Is Accepable

An IGRA is preerred or testing persons rom groups thathistorically have low rates o returning to have Ss readFor example, use o an IGRA might increase test completion rates or homeless persons and drug-users. Te useo IGRAs or such persons can increase test completion

rates, so control eorts can ocus on those most likely tobenet rom urther evaluation and treatment.An IGRA is preerred or testing persons who have receivedBCG (as a vaccine or or cancer therapy). Use o IGRAs inthis population is expected to increase diagnostic specicityand improve acceptance o treatment or LBI.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

13/28

Vol. 59 / RR-5 Recommendations and Reports 11

Siuais i Which a tSt Is Preerred Bu aIGRA Is Accepable

A S is preerred or testing children aged

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

14/28

12 MMWR June 25, 2010

initial test are unusual, such as when the Nil value is higherthan typical or the population being tested (e.g., IFN-concentration or Nil by QF-G or QF-GI >0.7 IU/mlor most o the U.S. populations), the Nil value is appre-ciably greater than the value obtained withM. tuberculosisantigen stimulation (e.g. when IFN- concentration or

Nil by QF-G is 0.35 IU/ml greater than the concentra-tion obtained with either ESA-6 or CFP-10 stimulation,or when the number o spots or Nil by -Spot is ourspots greater than the number with either ESA-6 orCFP-10 stimulation), or the Mitogen value is lower than isexpected or the population being tested (e.g., the MitogenResponse by QF-G or QF-GI is

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

15/28

Vol. 59 / RR-5 Recommendations and Reports 13

and limitations o IGRAs in situations o importance to medi-cal care or tuberculosis control. Questions to address includethe ollowing (not listed in any order o priority):

Are IGRAs better at predicting subsequent active tuber-culosis than S?Are persons with discordant S and IGRA results at

increased risk or active tuberculosis compared with per-sons with concordant negative results?Are higher IFN- responses associated with a greater riskor developing active tuberculosis?Do IGRAs perorm dierently in children than in adults,in those with extrapulmonary versus pulmonary tuberculo-sis, in those with HIV inection versus those without HIVinection, in those recently inected as compared with thoseinected years earlier, and in those with latent inection ascompared with those with active tuberculosis?Why do simultaneously perormed S, QF-GI, QF-G, and -Spot results dier?Can sensitivity and specicity o IGRAs be improved bymodication in testing methods, application o dierentinterpretation criteria, or inclusion o additional antigens?What is the best approach or determining cut points orIGRA interpretation, including situations where Nil valuesare high or Mitogen values are low?o what extent does inclusion o a borderline interpreta-tion improve IGRA accuracy?What causes variation in IGRA results and to what extent?What magnitude o change in IFN- response indicatesnew inection?

Ater exposure, how long does it take or an IGRA to

become positive?What is the clinical signicance o IGRA reversion?What methods should be used to monitor IGRA quality?Is there an association between lymphocyte count andIFN- response (with or without HIV inection)?What eect does treatment o M. tuberculosisinectionhave on IGRA results?How do host and bacterial genetic actors aect IGRAresults?

Reereces1. American Toracic Society, CDC. Diagnostic standards and classica-

tion o tuberculosis in adults and children. Am J Respir Crit Care Med2000;161:137695.

2. Rothel JS, Jones SL, Corner LA, Cox JC, Wood PR. A sandwich enzymeimmunoassay or bovine intereron-gamma and its use or the detectiono tuberculosis in cattle. Aust Vet J 1990:67;1347.

3. Converse PJ, Jones SL, Astemborski J, Vlahov D, Graham NM.Comparison o a tuberculin intereron-gamma assay with the tuberculinskin test in high-risk adults: eect o human immunodeciency virusinection. J Inect Dis 1997;176:144150.

4. Streeton JA, Desem N, Jones SL. Sensitivity and specicity o a gammaintereron blood test or tuberculosis inection. Int J uberc Lung Dis1998;2:44350.

5. Food and Drug Administration. QuantiFERON-B - P010033. Availableat http://www.da.gov/MedicalDevices/ProductsandMedicalProceduresDeviceApprovalsandClearances/Recently-ApprovedDevicesucm084025.htm. Accessed June 16, 2010.

6. Mazurek GH, Villarino ME. Guidelines or using the QuantiFERONB test or diagnosing latent Mycobacterium tuberculosisinectionMMWR 2003;52(No. RR-2):158.

7. Food and Drug Administration. QuantiFERON-B Gold P010033/S006. Available at http://www.da.gov/MedicalDevicesProductsandMedicalProcedures/DeviceApprovalsandClearances/PMAApprovals/ucm110838.htm. Accessed June 16, 2010.

8. CDC. Guidelines or using the QuantiFERON-B Gold test odetectingMycobacterium tuberculosis inection, United States. MMWR2005;54:(No. RR-15):4955.

9. World Health Organization. Global tuberculosis control: epidemiologystrategy, nancing: WHO report 2009. Geneva, Switzerland: WorldHealth Organization; 2009. Available at http://www.who.int/tb/publcations/global_report/2009/pd/report_without_annexes.pd. Accessed

June 16, 2010.10. CDC. Reported tuberculosis in the United States, 2008. Atlanta, GA

U.S. Department o Health and Human Services, CDC; 2009.11. Bennett DE, Courval JM, Onorato I, et al. Prevalence o tuberculosi

inection in the United States population: the national health and

nutrition examination survey, 19992000. Am J Respir Crit Care Med2008;177:34855.

12. CDC. argeted tuberculin testing and treatment o latent tuberculosinection. MMWR 2000;49:(No. RR-6).

13. Edwards PQ, Edwards LB. Story o the tuberculin skin test rom anepidemiologic viewpoint. Am Rev Respir Dis 1960;81:147.

14. Antonucci G, Girardi E, Raviglione MC, Ippolito G. Risk actors ortuberculosis in HIV-inected persons. JAMA 1995;274:1438.

15. Selwyn PA, Hartel D, Lewis VA, et al. A prospective study o the risko tuberculosis among intravenous drug users with human immunodeciency virus inection. N Engl J Med 1989;320:54550.

16. Horsburgh CR, Jr. Priorities or the treatment o latent tuberculosiinection in the United States. N Engl J Med 2004;350:20607.

17. Judson FN, Feldman RA. Mycobacterial skin tests in humans 12 yearater inection with Mycobacterium marinum. Am Rev Respir Di

1974;109:5447.18. Snider DE Jr. Bacille Calmette-Guerin vaccinations and tuberculin skintests. JAMA 1985;253:343839.

19. Mazurek GH, Weis SE, Moonan PK, et al. Prospective comparison otuberculin skin test and two whole blood intereron-gamma release assayin tuberculosis suspects. Clin Inect Dis 2007;45:83745.

20. Andersen P, Munk ME, Pollock JM, Doherty M. Specic immunebased diagnosis o tuberculosis. Lancet 2000;356:1099104.

21. Cellestis Limited. QuantiFERON-B Gold [Package insert]. Available ahttp://www.cellestis.com/IRM/Company/ShowPage.aspx?CPID=1247

Accessed June 16, 2010.22. Food and Drug Administration. QuantiFERON-B Gold In-ube

- P010033/S011. Available at http://www.da.gov/MedicalDevices/ProductsandMedicalProcedures/DeviceApprovalsandClearances/PMAApprovals/ucm106548.htm. Accessed June 16, 2010.

23. Food and Drug Administration. -SPO-B - P070006. Available athttp://www.da.gov/MedicalDevices/ProductsandMedicalProceduresDeviceApprovalsandClearances/PMAApprovals/ucm102794.htm

Accessed June 16, 2010.24. Oxord Immunotec Limited. -Spot.B [U.S. package insert]. Available a

http://www.oxordimmunotec.com/USpageInsert . Accessed June 16, 201025. Oxord Immunotec Limited. -Spot.B [U.K. package insert]. Availabl

at http://www.oxordimmunotec.com/96-UK. Accessed June 16, 2010.26. Harada N, Higuchi K, Yoshiyama , et al. Comparison o the sensitiv

ity and specicity o two whole blood intereron-gamma assays or Mtuberculosis inection. J Inect 2008;56:34853.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

16/28

14 MMWR June 25, 2010

27. siouris SJ, Coetzee D, oro PL, et al. Sensitivity analysis and potentialuses o a novel gamma intereron release assay or diagnosis o tubercu-losis. J Clin Microbiol 2006;44:284450.

28. Detjen AK, Keil , Roll S, et al. Intereron-gamma release assays improvethe diagnosis o tuberculosis and nontuberculous mycobacterial diseasein children in a country with a low incidence o tuberculosis. Clin InectDis 2007;45:3228.

29. Pai M, Joshi R, Bandyopadhyay M, et al. Sensitivity o a whole-bloodintereron-gamma assay among patients with pulmonary tuberculosisand variations in -cell responses during anti-tuberculosis treatment.Inection 2007;35:98103.

30. Adetia IM, Lugos MD, Hammond A, et al. Comparison o two interer-on gamma release assays in the diagnosis o Mycobacterium tuberculosisinection and disease in Te Gambia. BMC Inect Dis 2007;7:122.

31. Dominguez J, Ruiz-Manzano J, De Souza-Galvao M, et al. Comparisono two commercially available gamma intereron blood tests or immu-nodiagnosis o tuberculosis. Clin Vaccine Immunol 2008;15:16871.

32. Palazzo R, Spensieri F, Massari M, et al. Use o whole-blood samples inin-house bulk and single-cell antigen-specic gamma intereron assaysor surveillance oMycobacterium tuberculosis inections. Clin VaccineImmunol 2008;15:32737.

33. Chee CB, Gan SH, KhinMar KW, et al. Comparison o sensitivies o twocommercial gamma intereron release assays or pulmonary tuberculosis.

J Clin Microbiol 2008;46:193540.34. Ruhwald M, Bodmer , Maier C, et al. Evaluating the potential o IP-10

and MCP-2 as biomarkers or the diagnosis o tuberculosis. Eur RespirJ 2008;32:160715.

35. Bartu V, Havelkova M, Kopecka E. QuantiFERON-B Gold in thediagnosis o active tuberculosis. J Int Med Res 2008;36:43437.

36. Raby E, Moyo M, Devendra A, et al. Te eects o HIV on the sensitiv-ity o a whole blood IFN-gamma release assay in Zambian adults withactive tuberculosis. PLoS ONE 2008;3:e2489.

37. Aichelburg MC, Rieger A, Breitenecker F, et al. Detection and predictiono active tuberculosis disease by a whole-blood intereron-gamma releaseassay in HIV-1-inected individuals. Clin Inect Dis 2009;48:95462.

38. Goletti D, Steania C, Butera O, et al. Accuracy o immunodiagnostictests or active tuberculosis using single and combined results: a multi-center BNE-Study. PLoS ONE 2008;3:e3417.

39. Kampmann B, Whittaker E, Williams A, et al. Intereron-gamma releaseassays do not identiy more children with active tuberculosis than thetuberculin skin test. Eur Respir J 2009;33:137482.

40. Lee JY, Choi HJ, Park IN, et al. Comparison o two commercial inter-eron-gamma assays or diagnosingMycobacterium tuberculosisinection.Eur Respir J 2006;28:2430.

41. Meier , Eulenbruch HP, Wrighton-Smith P, Enders G, Regnath .Sensitivity o a new commercial enzyme-linked immunospot assay (SPO-B) or diagnosis o tuberculosis in clinical practice. Eur J ClinMicrobiol Inect Dis 2005;24:52936.

42. Goletti D, Carrara S, Vincenti D, et al. Accuracy o an immune diag-nostic assay based on RD1 selected epitopes or active tuberculosis in aclinical setting: a pilot study. Clin Microbiol Inect 2006;12:54450.

43. Ferrara G, Losi M, DAmico R, et al. Use in routine clinical practice otwo commercial blood tests or diagnosis o inection withMycobacteriumtuberculosis: a prospective study. Lancet 2006;367:132834.

44. Jaari C, Ernst M, Kalsdor B, et al. Rapid diagnosis o smear-negativetuberculosis by bronchoalveolar lavage enzyme-linked immunospot. Am

J Respir Crit Care Med 2006;174:104854.45. Kang YA, Lee HW, Hwang SS, et al. Useulness o whole-blood

intereron-gamma assay and intereron-gamma enzyme-linked immu-nospot assay in the diagnosis o active pulmonary tuberculosis. Chest2007;132:95965.

46. Bosshard V, Roux-Lombard P, Perneger , et al. Do results o the-SPO.B intereron-gamma release assay change ater treatment otuberculosis? Respir Med 2009;103:304.

47. Wang JY, Chou CH, Lee LN, et al. Diagnosis o tuberculosis by anenzyme-linked immunospot assay or intereron-gamma. Emerg InecDis 2007;13:5538.

48. Janssens JP, Roux-Lombard P, Perneger , et al. Quantitative scoring oan intereron-gamma assay or dierentiating active rom latent tuberculosis. Eur Respir J 2007;30:7228.

49. Ozekinci , Ozbek E, Celik Y. Comparison o tuberculin skin test anda specic -cell-based test, -Spot.B, or the diagnosis o latent tuberculosis inection. J Int Med Res 2007;35:696703.

50. Soysal A, orun , Ee S, et al. Evaluation o cut-o values o intererongamma-based assays in the diagnosis oM. tuberculosisinection. Int Juberc Lung Dis 2008;12:506.

51. Liao CH, Chou CH, Lai CC, et al. Diagnostic perormance o anenzyme-linked immunospot assay or intereron-gamma in extrapulmonary tuberculosis varies between dierent sites o disease. J Inec2009;59:4028.

52. Higuchi K, Kawabe Y, Mitarai S, Yoshiyama , Harada N, Mori Comparison o perormance in two diagnostic methods or tuberculosiinection. Med Microbiol Immunol 2009:198;337.

53. Adams LV, Waddell RD, von Reyn CF. -SPO.B est(R) results inadults withMycobacterium avium complex pulmonary disease. Scand JInect Dis 2008;40:196203.

54. Pai M, Zwerling A, Menzies D. Systematic review: -cell-based assay

or the diagnosis o latent tuberculosis inection: an update. Ann InternMed 2008;149:17784.

55. Farhat M, Greenaway C, Pai M, Menzies D. False-positive tuberculinskin tests: what is the absolute eect o BCG and non-tuberculoumycobacteria? Int J uberc Lung Dis 2006;10:1192204.

56. Wang L, urner MO, Elwood RK, Schulzer M, FitzGerald JM. Ameta-analysis o the eect o Bacille Calmette-Guerin vaccination ontuberculin skin test measurements. Torax 2002;57:8049.

57. Brock I, Weldingh K, Lillebaek , Follmann F, Andersen P. Comparisono tuberculin skin test and new specic blood test in tuberculosis contacts

Am J Respir Crit Care Med 2004;170:659.58. Arend SM, Tijsen SF, Leyten EM, et al. Comparison o two intereron

gamma assays and tuberculin skin test or tracing tuberculosis contactsAm J Respir Crit Care Med 2007;175:61827.

59. Choi JC, Shin JW, Kim JY, et al. Te eect o previous tuberculin

skin test on the ollow-up examination o whole-blood intererongamma assay in the screening or latent tuberculosis inection. Ches2008;133:141520.

60. Leung CC, Yam WC, Yew WW, et al. Comparison o -Spot.B and tuberculin skin test among silicotic patients. Eur Respir J 2008;31:26672.

61. Diel R, Loddenkemper R, Meywald-Walter K, Gottschalk R, NienhauA. Comparative perormance o tuberculin skin test, QuantiFERONB-Gold In ube assay, and -Spot.B test in contact investigations otuberculosis. Chest 2009;135:10108.

62. Hill PC, Jeries DJ, Brookes RH, et al. Using ELISPO to exposealse positive skin test conversion in tuberculosis contacts. PLoS ONE2007;2:e183.

63. Igari H, Watanabe A, Sato . Booster phenomenon o QuantiFERONB Gold ater prior intradermal PPD injection. Int J uberc Lung Di2007;11:78891.

64. Leyten EM, Prins C, Bossink AW, et al. Eect o tuberculin skin testingon a Mycobacterium tuberculosis-specic intereron-gamma assay. EuRespir J 2007;29:12126.

65. Naseer A, Naqvi S, Kampmann B. Evidence or boostingMycobacteriumtuberculosis-specic IFN-gamma responses at 6 weeks ollowing tuberculin skin testing. Eur Respir J 2007;29:128283.

66. Richeldi L, Ewer K, Losi M, et al. Repeated tuberculin testing does noinduce alse positive ELISPO results. Torax 2006;61:180.

67. Hill PC, Brookes RH, Fox A, et al. Longitudinal assessment o anELISPO test or Mycobacterium tuberculosisinection. PLoS Med2007;4:e192.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

17/28

Vol. 59 / RR-5 Recommendations and Reports 15

68. Ewer K, Millington KA, Deeks JJ, et al. Dynamic antigen-specic -cellresponses ater point-source exposure toMycobacterium tuberculosis. Am

J Respir Crit Care Med 2006;174:8319.69. Pai M, Joshi R, Dogra S, et al. Serial testing o health care workers or

tuberculosis using intereron-gamma assay. Am J Respir Crit Care Med2006;174:34955.

70. Perry S, Sanchez L, Yang S, et al. Reproducibility o QuantiFERON-Bgold in-tube assay. Clin Vaccine Immunol 2008;15:42532.

71. Veerapathran A, Joshi R, Goswami K, et al. -cell assays or tuberculosisinection: deriving cut-os or conversions using reproducibility data.PLoS ONE 2008;3:e1850.

72. siouris SJ, Austin J, oro P, et al. Results o a tuberculosis-specicIFN-gamma assay in children at high risk or tuberculosis inection. Int

J uberc Lung Dis 2006;10:93941.73. Nakaoka H, Lawson L, Squire SB, et al. Risk or tuberculosis among

children. Emerg Inect Dis 2006;12:138388.74. Janssens J, Roux-Lombard P, Perneger , et al. Contribution o a IFN-

gamma assay in contact tracing or tuberculosis in a low-incidence, highimmigration area. Swiss Med Wkly 2008;138:58593.

75. Vynnycky E, Fine PE. Lietime risks, incubation period, and serialinterval o tuberculosis. Am J Epidemiol 2000;152:24763.

76. Diel R, Loddenkemper R, Meywald-Walter K, Niemann S, Nienhaus A.Predictive value o a whole blood IFN-gamma assay or the development

o active tuberculosis disease ater recent inection withMycobacteriumtuberculosis. Am J Respir Crit Care Med 2008;177:116470.

77. Stout JE, Menzies D. Predicting tuberculosis: does the IGRA tell thetale? Am J Respir Crit Care Med 2008;177:10557.

78. Kik SV, Franken WP, Mensen M, et al. Predictive value or pro-gression to tuberculosis by IGRA and S in immigrant contacts.Eur Respir J 2009. Available at http://erj.ersjournals.com/cgi/rapidpd/09031936.00098509v1 . Accessed June 16, 2010.

79. Hill PC, Jackson-Sillah DJ, Fox A, et al. Incidence o tuberculosis andthe predictive value o ELISPO and Mantoux tests in Gambian casecontacts. PLoS ONE 2008;3:e1379.

80. Newton SM, Brent AJ, Anderson S, Whittaker E, Kampmann B.Paediatric tuberculosis. Lancet Inect Dis 2008;8:498510.

81. Kampmann B, ena-Coki G, Anderson S. Blood tests or diagnosis otuberculosis. Lancet 2006;368:2823.

82. Connell G, Curtis N, Ranganathan SC, Buttery JP. Perormance o awhole blood intereron gamma assay or detecting latent inection withMycobacterium tuberculosis in children. Torax 2006;61:61620.

83. Lewinsohn DA, Zalwango S, Stein CM, et al. Whole blood intereron-gamma responses to Mycobacterium tuberculosis antigens in younghousehold contacts o persons with tuberculosis in Uganda. PLoS ONE2008;3:e3407.

84. Bergamini BM, Losi M, Vaienti F, et al. Perormance o commercialblood tests or the diagnosis o latent tuberculosis inection in childrenand adolescents. Pediatrics 2009;123:e41924.

85. Connell G, Ritz N, Paxton GA, et al. A three-way comparison otuberculin skin testing, QuantiFERON-B gold and -SPO.B inchildren. PLoS ONE 2008;3:e2624.

86. Dogra S, Narang P, Mendiratta DK, et al. Comparison o a whole bloodintereron-gamma assay with tuberculin skin testing or the detectiono tuberculosis inection in hospitalized children in rural India. J Inect2007;54:26776.

87. Lighter J, Rigaud M, Eduardo R, Peng CH, Pollack H. Latent tuberculo-sis diagnosis in children by using the QuantiFERON-B Gold In-ubetest. Pediatrics 2009;123:307.

88. Nicol MP, Davies MA, Wood K, et al. Comparison o -SPO.B assayand tuberculin skin test or the evaluation o young children at high riskor tuberculosis in a community setting. Pediatrics 2009;123:3843.

89. Warier A, Gunawathi S, Sankarapandian V, John KR, Bose A. -cell assayas a diagnostic tool or tuberculosis. Indian Pediatr 2010;47:902.

90. aylor RE, Cant AJ, Clark JE. Potential eect o NICE tuberculosis guidelines on paediatric tuberculosis screening. Arch Dis Child2008;93:2003.

91. Shingadia D, Novelli V. Te tuberculin skin test: a hundred, not outArch Dis Child 2008;93:18990.

92. Brock I, Ruhwald M, Lundgren B, et al. Latent tuberculosis in HIVpositive, diagnosed by the M. tuberculosisspecic intereron gammatest. Respir Res 2006;7:56.

93. Balcells ME, Perez CM, Chanqueo L, et al. A comparative study o twodierent methods or the detection o latent tuberculosis in HIV-positivindividuals in Chile. Int J Inect Dis 2008;12:64552.

94. Jones S, de Gijsel D, Wallach FR, et al. Utility o QuantiFERON-BGold in-tube testing or latent B inection in HIV-inected individualsInt J uberc Lung Dis 2007;11:11905.

95. Luetkemeyer AF, Charlebois ED, Flores LL, et al. Comparison o anintereron-gamma release assay with tuberculin skin testing in HIVinected individuals. Am J Respir Crit Care Med 2007;175:73742.

96. alati NJ, Seybold U, Humphrey B, Aina A, apia J, et al. Poor con-cordance between intereron-gamma release assays and tuberculin skintests in diagnosis o latent tuberculosis inection among HIV-inectedindividuals. BMC Inect Dis 2009;9:15.

97. Bocchino M, Matarese A, Belloore B, et al. Perormance o two commercial blood IFN-gamma release assays or the detection oMycobacterium

tuberculosisinection in patient candidates or anti-NF-alpha treatmentEur J Clin Microbiol Inect Dis 2008;27:90713.

98. Cobanoglu N, Ozcelik U, Kalyoncu U, et al. Intereron-gamma assayor the diagnosis o tuberculosis inection beore using tumour necrosiactor-alpha blockers. Int J uberc Lung Dis 2007;11:117782.

99. Matulis G, Juni P, Villiger PM, Gadola SD. Detection o latent tuberculosis in immunosuppressed patients with autoimmune diseasesperormance o aMycobacterium tuberculosisantigen-specic intererongamma assay. Ann Rheum Dis 2008;67:8490.

100. Ponce de LD, Acevedo-Vasquez E, Alvizuri S, et al. Comparison o anintereron-gamma assay with tuberculin skin testing or detection otuberculosis (B) inection in patients with rheumatoid arthritis in B-endemic population. J Rheumatol 2008;35:77681.

101. Mandalakas AM, Hesseling AC, Chegou NN, et al. High level o

discordant IGRA results in HIV-inected adults and children. Int uberc Lung Dis 2008;12:41723.

102. Rangaka MX, Wilkinson KA, Seldon R, et al. Eect o HIV-1 inectionon -cell-based and skin test detection o tuberculosis inection. Am Respir Crit Care Med 2007;175:51420.

103. Stephan C, Wol , Goetsch U, et al. Comparing QuantiFERONtuberculosis gold, -SPO tuberculosis and tuberculin skin test inHIV-inected individuals rom a low prevalence tuberculosis country

AIDS 2008;22:24719.104. Lindemann M, Dioury Y, Beckebaum S, et al. Diagnosis o tuberculosi

inection in patients awaiting liver transplantation. Hum Immuno2009;70:248.

105. Passalent L, Khan K, Richardson R, et al. Detecting latent tuberculosiinection in hemodialysis patients: a head-to-head comparison o th

-SPO.B test, tuberculin skin test, and an expert physician panelClin J Am Soc Nephrol 2007;2:6873.

106. Piana F, Codecasa LR, Cavallerio P, et al. Use o a -cell-based tesor detection o tuberculosis inection among immunocompromisedpatients. Eur Respir J 2006;28:314.

107. Porsa E, Cheng L, Graviss EA. Comparison o an ESA-6/CFP-10peptide-based enzyme-linked immunospot assay to a tuberculin skintest or screening o a population at moderate risk o contractingtuberculosis. Clin Vaccine Immunol 2007;14:7149.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

18/28

16 MMWR June 25, 2010

108. Vassilopoulos D, Stamoulis N, Hadziyannis E, Archimandritis AJ.Useulness o enzyme-linked immunospot assay (elispot) comparedto tuberculin skin testing or latent tuberculosis screening in rheu-matic patients scheduled or anti-tumor necrosis actor treatment. JRheumatol 2008;35:12716.

109. MAG Mutual Healthcare Solutions, Inc. Physicians ee and codingguide 2009. Atlanta, Georgia: MAG Mutual Healthcare Solutions,

Inc.; 2009.110. Marra F, Marra CA, Sadatsaavi M, et al. Cost-eectiveness o a new

intereron-based blood assay, QuantiFERON(R)-B Gold, in screeningtuberculosis contacts. Int J uberc Lung Dis 2008;12:141424.

TABLE 1. Interpretation criteria or the QuantiFERON-TB GoldTest (QFT-G)

Interpretation Nil* TB ResponseMitogen

Response

Positive Any 0.35 IU/ml and 50% o Nil AnyNegative** 0.7

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

19/28

Vol. 59 / RR-5 Recommendations and Reports 17

TABLE 4. QuantiFERON-TB Gold-In-Tube test (QFT-GIT) sensitivity,* by country in which study was conducted 14 countries20062009

Confrmed TB QFT-GIT results TST results

Country Subjects

No.

confrmed/

No. with TB

diagnosis (%)

HIV-positive

Inter-

pretation

criteria**

Positive Indeterminate

Cuto

Positive % TST+

vs.

QFT-GIT

p-valueNo. +/

No. tested (%)

No. +/

No. valid (%)

No. +/

No. tested (%)

No. +/

No. tested (%)

South Arica Children 154/154 (100) 26/41 (63) A 100/131 (76) 23/154 (15) Stratifed 131/146 (90)

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

20/28

18 MMWR June 25, 2010

TABLE 5. T-SPOT.TB test (T-Spot) sensitivity,* by country in which study was conducted 12 countries, 20052009

Confrmed TB T-Spot results TST results

Country Subjects

No.

confrmed/

No. with TB

diagnosis (%)

HIV-positiveInter-

pretation

criteria**

Positive Indeterminate

Cuto

Positive % TST+

vs.

QFT-GIT+

p-valueNo. +/

No. tested (%)

No. +/

No. valid (%)

No. +/

No. tested (%)

No. +/

No. tested (%)

Singapore Adults 286/286 (100) 7/238 (3) A 254/270 (94) 3/286 (1) 10 mm 206/217 (95) 0.84

15 mm 158/217 (73) ND

Spain*** Adults & children NR/42 (NR) NR NR B 36/39 (86) 3/42 (7) 5 mm 40/42 (95) 0.93

Germany Children aged07 yrs

28/28 (100) NR NR B 26/28 (93) 0/28 (0) 5 mm 28/28 (100) 0.49

South Korea Adults 37/65 (57) 0/31 (0) C 83/87 (95) 0/87 (0) 5 mm 64/87 (74) 15 yrs

58/67 (87) 0/67 (0) H 59/64 (92) 3/67 (4) 10 mm 45/66 (68) 15 yrs

58/58 (100) 0/58 (0) K 57/58 (98) 0/58 (2) ND ND ND ND

Turkey****** Adults NR/28 NR/28 NR NR B 26/28 (93) NR NR 10 mm 23/28 (82) 0.42

Turkey Adults & childrenaged >15 yrs

100/100 (100) 0/100 (0) L 80/96 (83) 4/100 (4) 10 mm 80/99 (81) 0.79

MultipleEuropean

Adults 69/69 (100) 3/NR (NR) B 62/69 (90) 0/69 (0) 10 or 15 114/136 (84) 0.06

0/19 (0) 0/NR (NR) 13/19 (68) 0/19 (0) 37/41 (90) 0.09

Taiwan Adults with extra-pulmonary TB

50/50 (100) 2/NR (NR) M 40/50 (80) NR NR ND ND ND ND

0/39 (0) 31/39 (79)

UnitedKingdom*******

Children 25/25 (100) 0/35 (0) F 14/24 (58) 1/25 (8) 10 mm 21/24 (86) 0.05

0/38 (0) 17/34 (50) 4/38 (11) 24/38 (63) 0.38

Japan Adults 49/49 (100) NR NR N 47/47 (100) 2/49 (4) ND ND ND ND

See Table 5 ootnotes on the ollowing page.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

21/28

Vol. 59 / RR-5 Recommendations and Reports 19

TABLE 5. (Continued) T-SPOT.TB (T-Spot) sensitivity* results, by country in which study was conducted 12 countries, 20052009

* Source: Modifed rom Pai M, Zwerling A, Menzies D. Systematic review: T-cell-based assays or the diagnosis o latent tuberculosis inection: an update. Ann Intern Med2008;149:17784 supplemented with additional inormation and compared with TST specifcity when available.

Tuberculosis. Confrmed by culture and/or nucleic acid amplifcation test. Human immunodefciency virus. Tuberculin skin test.

** A = T-Spot was interpreted as positive i a test well (with either early secretory antigenic target-6 [ESAT-6] or culture fltrate protein culture fltrate protein [CFP-10]) con-tained 6 spots or more than the negative control well and had at least twice the spots as the negative control well, and the negative control well had 10 spots; indeterminate i not positive and the mitogen control well had 10 spots. B = T-Spot interpretation criteria were not explicitly stated. C

= T-Spot was interpreted as positive i a test well (with either ESAT-6 or CFP-10) contained 5 spots or more than the negative control well and had at least twice the spotsas the negative control well and the negative control well had 10 spots and as indeterminate i the negative control well had >10 spots. D = T-Spot was interpreted aspositive i a test well (with either ESAT-6 or CFP-10) contained 5 spots or more than the negative control well and had at least twice the spots as the negative control welland the mitogen control well had >20 spots and indeterminate i the mitogen control well had 20 spots. E = T-Spot was interpreted as positive i the well with ESAT-6contained at least twice the average number o spots as the negative control well or the well with CFP-10 contained at least 4 times the average number o spots as thenegative control well. F = T-Spot was interpreted as positive i a test well (with either ESAT-6 or CFP-10) contained 6 spots or more than the negative control well andhad at least twice the spots as the negative control well and the negative control well had

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

22/28

20 MMWR June 25, 2010

TABLE 6. QuantiFERON-TB Gold In-Tube test (QFT-GIT) specifcity,* by country in which study was conducted our countries20072008

Country Subjects

BCG-vaccinated

HIV-positive

QFT-GIT results TST Results

% TST-vs. %

QFT-GIT-

p-value

Inter-pretation

criteria**

Negative Indeterminate

Cuto

Negative

No.vaccinated/

No.

evaluated (%)

No. +/

No. tested (%)

No. +/

No. valid (%)

No. +/

No. tested (%)

No. +/

No. tested (%)

Germany Children aged011 yrs w/lymphadenitis

0/23 (0) NR NR A 19/19 (100) ND*** ND 5 2/23 (9)

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

23/28

Vol. 59 / RR-5 Recommendations and Reports 21

TABLE 7. T-SPOT.TB test (T-Spot) specifcity,* by country in which study was conducted three countries, 20062008

Country Subjects

BCG-vaccinated

% TST-vs.

% T-Spot-p-value

T-Spot results TST results

No.vaccinated/

No.evaluated (%) HIV status

Inter-pretationcriteria**

Negative Indeterminate

Cuto

Negative

No. +/

No. valid (%)

No. +/

No. tested (%)

No. +/

No. tested (%)

Germany Children aged 011 yrs w/lymphadenitis

0/19 (0) NR A 18/19 (95) 4/23 (17) 5 2/23 (9)

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

24/28

22 MMWR June 25, 2010

TABLE 8. Summary o fndings o published studies evaluating QuantiFERON-TB Gold-In-Tube test (QFT-GIT) and/or T-SPOT.TBtest (T-Spot) in tuberculosis contacts compared with tuberculin skin test (TST) when available, by country in which study wasconducted seven countries, 20062008

Country Subjects

BCG vaccinated*

TSTcuto Findings

No. vacci-nated/No.evaluated (%)

South Arica Children aged 515 yrs 115/174 (66) 10 mm QFT-GIT and TST results were associated with older age but not with recentor remote household contact.

Nigeria Child contacts & controlsaged 114 yrs

187/207 (90) 10 mm QFT-GIT and TST results were associated with acid-ast bacillus (AFB)status o source and age or children living with AFB-negative persons andcontrols. +TST/-QFT-GIT discordance was more common in controls andchildren living with AFB-negative persons. -TST/+ QFT-GIT were more com-mon in children living with AFB-positive persons.

Denmark Adult contacts w/out BCG 0/785 (0) 10 mm TST results were associated with age but not with est imates o exposure.T-Spot results were associated with an estimate o exposure (cumulativeshopping time). QFT-GIT (without mitogen) was associated with cumulativeshopping time more so than T-Spot.

The Gambia** Adult & child contactsaged 15 yrs

84/194 (43) 10 mm TST more strongly associated with exposure gradient than QFT-GIT (withoumitogen). For contacts sleeping in the same room as compared with thosesleeping in dierent houses, the odds ratio or a positive TST was 4.8 (95%confdence interval [CI] = 1.317.1) as compared with 3.8 (CI = 1.212.5) or

QFT-GIT.

Switzerland Adult & child contactsaged 1683 yrs

238/295 (81) 10 mm Both TST & T-Spot results were associated with age, gender, BCG, andincidence o tuberculosis in country o origin, but not to any o 5 exposurescores.

Germany Adult & child contacts w/TST >5 mm

453/812 (56) NA Both QFT-GIT & T-Spot results were associated with age, AFB + or cough-ing source, cumulative exposure time, and oreign origin. Associations withTST results were not assessed.

Spain Adults & children 128/270 (47) 5 mm TST results were associated with BCG. QFT-GIT & T-Spot results were notassociated with BCG. Association o test results with incidence o tuberculo-sis in country o origin was not assessed.

* Bacillus Calmette-Guerin.Source: Tsiouris SJ, Austin J, Toro P et al. Results o a tuberculosis-specifc IFN-gamma assay in children at high risk or tuberculosis inection. Int J

Tuberc Lung Dis 2006;10:93941. Source: Nakaoka H, Lawson L, Squire SB, et al. Risk or tuberculosis among children. Emerg Inect Dis 2006;12:13838.

Source: Arend SM, Thijsen SF, Leyten EM, et al. Comparison o two intereron-gamma assays and tuberculin skin test or tracing tuberculosis contactsAm J Respir Crit Care Med 2007;175:61827.

** Source: Adetia IM, Lugos MD, Hammond A, et al. Comparison o two intereron gamma release assays in the diagnosis o Mycobacterium tuberculosisinection and disease in The Gambia. BMC Inect Dis 2007;7:122.

Source: Janssens J, Roux-Lombard P, Perneger T, Metzger M, Vivien R, Rochat T. Contribution o a IFN-gamma assay in contact tracing or tuberculosisin a low-incidence, high immigration area. Swiss Med Wkly 2008;138:58593.

Source: Diel R, Loddenkemper R, Meywald-Walter K, Gottschalk R, Nienhaus A. Comparative perormance o tuberculin skin test, QuantiFERON-TBGold In Tube assay, and T-Spot.TB test in contact investigations or tuberculosis. Chest 2009;135:10108.

Source: Dominguez J, Ruiz-Manzano J, De Souza-Galvao M, et al. Comparison o two commercially available gamma intereron blood tests or immu-nodiagnosis o tuberculosis. Clin Vaccine Immunol 2008;15:16871.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

25/28

Vol. 59 / RR-5 Recommendations and Reports 23

TABLE 9. QuantiFERON-TB Gold In-Tube (QFT-GIT) test results in immunosuppressed persons compared with tuberculin skintest (TST) results when available 10 countries, 20062008

QFT-GIT results TST results

% TST+vs. %

QFT-GIT+

p-value

Positive Indeterminate Positive

Country Subjects HIV* statusNo. +/

No. valid (%)No. +/

No. tested (%) CutoNo. +/

No. tested (%)

Denmark

607 adults 607 HIV+ 27/570 (4.7) 20/590 (3.4) ND

ND ND NDChile** 116 adults 116 HIV+ 17/115 (15) 0/115 (0) 5 mm 12/110 (11) 0.50United States 207 adults 207 HIV+ 11/191 (6) 10/201 (5) 5 mm 13/201 (7) 0.94

United States 294 adults 294 HIV+ 25/279 (9) 15/294 (5) 5 mm 19/205 (9) 0.99

Zambia 112 adults withsmear + TB

59 HIV+ 37/49 (76) 10/59 (17) 5 mm 26/47 (55) 0.0637 HIV- 31/32 (97) 5/37 (14) 25/31 (81) 0.09

16 not tested 15/15 (100) 1/16 (6) 0/14 (0)

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

26/28

24 MMWR June 25, 2010

TABLE 10. Published studies evaluating T-SPOT.TB test (T-Spot) among immunosuppressed persons compared with tuberculinskin test (TST) when available eight countries, 20062008

T-Spot results TST results

% TST+vs.

% T-Spot+p-value

Positive Indeterminate

Cuto

Positive

Country Subjects HIV* StatusNo. +/

No. valid (%)No. +/

No. tested (%)No. +/

No. tested (%)

South Arica

20 HIV+ adults 20 HIV+ 13/18 (72) 2/20 (10) 5 mm 10/16 (63) 0.8123 HIV+ children 23 HIV+ 12/23 (52) 0/23 (0) 6/23 (26) 0.13

South Arica 160 adults at HIV screening clinic 74 HIV+ 38/73 (52) 1/74 (1) 5 mm 35/67 (52) 0.99

86 HIV- 51/86 (59) 0/86 (0) 66/77 (86)

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

27/28

Vol. 59 / RR-5 Recommendations and Reports 25

IGRA Exper Cmmiee MembersMembership as Augus 2008

Chair: Neil Schluger, MD, Columbia University, New York, New York

Moderator:John Seggerson, Stop B USA, Atlanta, GA

Members: Paul Barnicott, U.S. Air Force School o Aerospace Medicine, San Antonio, exas; John Bernardo, MD, Boston University School o MedicineBoston, Massachusetts; Henry M. Blumberg, MD, Emory University School o Medicine, Atlanta, Georgia; Helene Calvet, MD, Long Beach Dept. o Healtand Human Services, Long Beach, Caliornia; Charles Daley, MD, National Jewish Medical and Research Center, Denver, Colorado; Susan Dorman, MD

Johns Hopkins University School o Medicine, Baltimore, Maryland; Edward Graviss, PhD, Baylor College o Medicine, Houston, exas; iany HarrisPhD, New York City Dept. o Health and Mental Hygiene, New York, New York; Philip Hill, MD, University o Otago School o Medicine, Dunedin, NewZealand; Masae Kawamura, MD, San Francisco Department o Public Health, San Francisco, Caliornia; Lisa Keep, MD, Uniormed Services Univ. o thHealth Sciences, Bethesda, Maryland; Stephen Kralovic, MD, Cincinnati VA Medical Center, Cincinnati, Ohio; Michael Leonard, MD, Georgia Department oHuman Resources, Atlanta, Georgia; David Lewinsohn, MD, PhD, Oregon Health and Sciences University, Portland VA Medical Center, Deborah LewinsohnMD, Oregon Health and Sciences University, Portland, Oregon; Kathleen Moser, MD, San Diego County Department o Health, Poway, Caliornia; EdwardNardell, MD, Brigham and Womens Hospital, Boston, Massachusetts; Masa Narita, MD, Seattle and King County Public Health, Seattle, WashingtonRichard OBrien, MD, Foundation or Innovative New Diagnostics, Geneva, Switzerland; Randall Reves, MD, Denver Public Health Department, DenverColorado; Luca Richeldi, MD, PhD, University o Modena and Reggio Emilia, Modena, Italy; Kim Connelly Smith, MD, University o exas Health ScienceCenter, Jeery Starke, MD, exas Childrens Hospital, Baylor College o Medicine, Houston, exas; David Warshauer, PhD, Wisconsin State Laboratory oHygiene, Madison, Wisconsin; Gail Woods, MD, Central Arkansas Veterans Healthcare System, Little Rock, Arkansas.

IGRA Exper Cmmiee PreseersMembership as Augus 2008

Members: Sandra Arend, MD, PhD, Leiden University Medical Center, Leiden, Te Netherlands; John Bernardo, MD, Boston University School o MedicineBoston, Massachusetts; Henry M.Blumberg, MD, Emory University School o Medicine, Atlanta, Georgia; Charles Daley, MD, National Jewish Medicaand Research Center, Denver, Colorado; Roland Diel, MD, University o Dsseldor, School o Public Health, Dsseldor, Germany; Edward Graviss, MD,Baylor College o Medicine, Houston, exas; iany Harris, PhD, New York City Dept. o Health and Mental Hygiene, New York, New York; AnthonyHawkridge, MD, Aeras Global B Vaccine Foundation, Cape own, South Arica; Philip Hill, MD, University o Otago School o Medicine, Dunedin, NewZealand; Masae Kawamura, MD, San Francisco Department o Public Health, San Francisco, Caliornia; Deborah Lewinsohn, MD, Portland VA MedicaCenter, David Lewinsohn, MD, PhD, Oregon Health and Sciences University, Portland, Oregon; Hassan Mahomed, Mmed, University o Cape own, Capeown, South Arica; Freddie Poole, MS, Center or Devices and Radiological Health, Food and Drug Administration, Rockville, Maryland; Luca RicheldiMD, PhD, University o Modena and Reggio Emilia, Modena, Italy; James Rothel, PhD, Cellestis Limited, Carnegie, Victoria, Australia; Neil SchlugerMD, Columbia University, New York, New York; John Seggerson, SOP B USA, Atlanta, Georgia; Kim Connelly Smith, MD, University o exas HealthScience Center, Houston, exas; Peter Wrighton-Smith, DPhil, Oxord Immunotec, Inc., Oxord, United Kingdom; Jean-Pierre Zellweger, MD, Swiss Lung

Association, Lausanne, Switzerland; Kenneth Castro, MD, John Jereb, MD, Gerald Mazurek, MD, CDC, Atlanta, Georgia.

8/6/2019 A6 MMWR Update on IGRA in Diagnosis of TB 2010

28/28

U.S. Government Printing Oce: 2010-623-026/41257 Region IV ISSN: 1057-5987

MMWR

TeMorbidity and Mortality Weekly Report (MMWR) Series is prepared by the Centers or Disease Control and Prevention (CDC) and is available ree ocharge in electronic ormat. o receive an electronic copy each week, visitMMWRs ree subscription page at http://www.cdc.gov/mmwr/mmwrsubscribe.html.Paper copy subscriptions are available through the Superintendent o Documents, U.S. Government Printing Oce, Washington, DC 20402; telephone202-512-1800.

Data presented by the Notiable Disease Data eam and 122 Cities Mortality Data eam in the weeklyMMWRare provisional, based on weekly reports to

CDC by state health departments. Address all inquiries about theMMWRSeries, including material to be considered or publication, to Editor,MMWRSeries, Mailstop E-90, CDC, 1600 Cliton Rd., N.E., Atlanta, GA 30333 or to [email protected].

All material in theMMWRSeries is in the public domain and may be used and reprinted without permission; citation as to source, however, is appreciated.

Use o trade names and commercial sources is or identication only and does not imply endorsement by the U.S. Department o Health and Human