AMERICAN THORACIC SOCIETYDOCUMENTS

Treatment of Drug-Resistant TuberculosisAn Official ATS/CDC/ERS/IDSA Clinical Practice GuidelinePayam Nahid, Sundari R. Mase, Giovanni Battista Migliori, Giovanni Sotgiu, Graham H. Bothamley, Jan L. Brozek,Adithya Cattamanchi, J. Peter Cegielski, Lisa Chen, Charles L. Daley, Tracy L. Dalton, Raquel Duarte,Federica Fregonese, C. Robert Horsburgh, Jr., Faiz Ahmad Khan, Fayez Kheir, Zhiyi Lan, Alfred Lardizabal,Michael Lauzardo, Joan M. Mangan, Suzanne M. Marks, Lindsay McKenna, Dick Menzies, Carole D. Mitnick,Diana M. Nilsen, Farah Parvez, Charles A. Peloquin, Ann Raftery, H. Simon Schaaf, Neha S. Shah, Jeffrey R. Starke,John W. Wilson, Jonathan M. Wortham, Terence Chorba, and Barbara Seaworth; on behalf of the American ThoracicSociety, U.S. Centers for Disease Control and Prevention, European Respiratory Society, and Infectious DiseasesSociety of America

THIS OFFICIAL CLINICAL PRACTICE GUIDELINE WAS APPROVED BY THE AMERICAN THORACIC SOCIETY, THE EUROPEAN RESPIRATORY SOCIETY, AND THE INFECTIOUSDISEASES SOCIETY OF AMERICA SEPTEMBER 2019, AND WAS CLEARED BY THE U.S. CENTERS FOR DISEASE CONTROL AND PREVENTION SEPTEMBER 2019

Background: The American Thoracic Society, U.S. Centers forDisease Control and Prevention, European Respiratory Society, andInfectious Diseases Society of America jointly sponsored this newpractice guideline on the treatment of drug-resistant tuberculosis(DR-TB). The document includes recommendations on thetreatment of multidrug-resistant TB (MDR-TB) as well asisoniazid-resistant but rifampin-susceptible TB.

Methods: Published systematic reviews, meta-analyses, and a newindividual patient data meta-analysis from 12,030 patients, in 50studies, across 25 countries with confirmed pulmonary rifampin-resistant TB were used for this guideline. Meta-analytic approachesincluded propensity score matching to reduce confounding. Eachrecommendationwas discussed by an expert committee, screened forconflicts of interest, according to the Grading of Recommendations,Assessment, Development, and Evaluation (GRADE) methodology.

Results: Twenty-one Population, Intervention, Comparator,and Outcomes questions were addressed, generating 25GRADE-based recommendations. Certainty in the evidence

was judged to be very low, because the data camefrom observational studies with significant loss to follow-upand imbalance in background regimens between comparatorgroups. Good practices in the management of MDR-TB aredescribed. On the basis of the evidence review, a clinical strategytool for building a treatment regimen for MDR-TB is alsoprovided.

Conclusions: New recommendations are made for the choice andnumber of drugs in a regimen, the duration of intensive andcontinuation phases, and the role of injectable drugs for MDR-TB.On the basis of these recommendations, an effective all-oralregimen for MDR-TB can be assembled. Recommendations are alsoprovided on the role of surgery in treatment of MDR-TB and fortreatment of contacts exposed to MDR-TB and treatment ofisoniazid-resistant TB.

Keywords: MDR-TB; tuberculosis; duration of treatment; drugtreatment; treatment monitoring

ORCID IDs: 0000-0003-2811-1311 (P.N.); 0000-0001-5363-0637 (S.R.M.); 0000-0002-2597-574X (G.B.M.); 0000-0002-1600-4474 (G.S.);0000-0002-7092-8547 (G.H.B.); 0000-0002-3122-0773 (J.B.); 0000-0002-6553-2601 (A.C.); 0000-0001-6804-0111 (L.C.); 0000-0003-3324-926X (C.L.D.);0000-0001-6838-7895 (C.R.H.); 0000-0003-0473-8734 (F.A.K.); 0000-0002-4192-5080 (F.K.); 0000-0001-5519-2474 (Z.L.); 0000-0003-3273-1097 (A.L.);0000-0002-7096-4185 (M.L.); 0000-0001-6770-086X (J.M.M.); 0000-0003-3024-1940 (S.M.M.); 0000-0002-4703-0835 (L.M.);0000-0002-3455-658X (C.D.M.); 0000-0003-1211-5043 (F.P.); 0000-0001-9002-7052 (C.A.P.); 0000-0001-5755-4133 (H.S.S.);0000-0001-7722-0958 (T.C.); 0000-0003-2922-4940 (B.S.).

Supported by the American Thoracic Society, the United States Centers for Disease Control and Prevention, the European Respiratory Society, and theInfectious Diseases Society of America.

An Executive Summary of this document is available at http://www.atsjournals.org/doi/suppl/10.1164/rccm.201909-1874ST.

You may print one copy of this document at no charge. However, if you require more than one copy, you must place a reprint order. Domestic reprint orders:amy.schriver@sheridan.com; international reprint orders: louisa.mott@springer.com.

Correspondence and requests for reprints should be addressed to Payam Nahid, M.D., M.P.H, Zuckerberg San Francisco General Hospital, Division of Pulmonaryand Critical Care Medicine, UCSF Center for Tuberculosis, 1001 Potrero Avenue, Building 5, Room 5K1, San Francisco, CA 94110. E-mail: pnahid@ucsf.edu.

This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.

Am J Respir Crit Care Med Vol 200, Iss 10, pp e93–e142, Nov 15, 2019

Copyright © 2019 by the American Thoracic Society

DOI: 10.1164/rccm.201909-1874ST

Internet address: www.atsjournals.org

American Thoracic Society Documents e93

ContentsOverview

Summary of Good PracticesSummary of Recommendations

IntroductionGood Practices for Treating DR-TB

Diagnosing TB and Identificationof Drug Resistance

Treatment and Monitoring of DR-TB

Infection Control and DR-TBCase Management for DR-TB

Treatment of MDR-TB, Number ofDrugs, and Duration of TreatmentPhases

Number of Drugs in the RegimenDuration of Intensive andContinuation Phases in TreatingMDR-TB

Drugs and Drug ClassesAmoxicillin/ClavulanateBedaquilineCarbapenems with ClavulanicAcid

ClofazimineCycloserineDelamanidEthambutol

Ethionamide and ProthionamideFluoroquinolones: Levofloxacin,Moxifloxacin, Ciprofloxacin, andOfloxacin

Injectables: Amikacin,Capreomycin, Kanamycin, andStreptomycin

LinezolidMacrolides: Azithromycin andClarithromycin

p-Aminosalicylic AcidPyrazinamide

Building a Treatment Regimen forMDR-TB

Role of Therapeutic Drug Monitoring inTreatment of MDR-TB

Shorter-Course, Standardized, 9- to12-Month Regimen for MDR-TB

Summary of the EvidenceBenefitsHarmsAdditional ConsiderationsConclusionsResearch Needs

Role of Surgery in MDR-TBSummary of the EvidenceBenefitsHarms

Additional ConsiderationsConclusionsResearch Needs

Treatment of Isoniazid-Resistant TBSummary of the EvidenceBenefitsHarmsAdditional ConsiderationsConclusionsResearch Needs

Treatment of MDR-TB in SpecialSituations

HIV InfectionChildrenPregnant Women

Treatment of Contacts Exposed toMDR-TB

Summary of the EvidenceBenefitsHarmsAdditional ConsiderationsConclusionsResearch Needs

Summary of Key Differencesbetween ATS/CDC/ERS/IDSA andWHO 2019 Consolidated Guidelineson Drug-Resistant TuberculosisTreatment

Overview

Treatment of tuberculosis (TB), regardlessof the results of drug susceptibility testing(DST), is focused on both curing theindividual patient and minimizing thetransmission of Mycobacterium tuberculosisto other persons. Thus, effective treatmentof TB has benefits for both the individualpatient and the community in which thepatient resides. However, notablecomplexities need to be addressed tosuccessfully treat disease resulting fromdrug-resistant M. tuberculosis isolatescompared with treatment of drug-susceptible TB disease, including additionalmolecular and phenotypic diagnostic teststo determine drug susceptibility; the use ofsecond-line drugs, which have toxicitiesthat increase harms that must be balancedwith their benefits; and prolongedtreatment durations. The newrecommendations provided in thisguideline are for the treatment ofdrug-resistant TB (DR-TB), includingmultidrug-resistant TB (MDR-TB) andisoniazid-resistant TB, and are intended tohelp providers identify the therapeutic

options associated with improved outcomes(i.e., greater treatment success, feweradverse events, and fewer deaths) and in thecontext of individual patient values andpreferences. Worthy of emphasis, thecommittee recommends that only drugs towhich the patient’s M. tuberculosis isolatehas documented, or high likelihood of,susceptibility be included in an effectivetreatment regimen, noted as an ungradedgood practice statement, and consistentwith ongoing stewardship efforts for theoptimal use of antibiotics (1). Drugs knownto be ineffective on the basis of in vitrogrowth-based or molecular DST shouldnot be used. The following alphabeticallylisted drugs and drug classes wereconsidered for inclusion in treatmentregimens: amoxicillin/clavulanate,bedaquiline, carbapenem withclavulanic acid, clofazimine, cycloserine,delamanid, ethambutol, ethionamide,fluoroquinolones, injectable agents,linezolid, macrolides, p-aminosalicylic acid,and pyrazinamide. Of note, pretomanid incombination with bedaquiline and linezolidwas recently approved by U.S. Food andDrug Administration (FDA) for the

treatment of a specific limited population ofadults with pulmonary extensively drug-resistant (XDR-TB) or treatment-intolerantor nonresponsive MDR-TB; however, thepreparation and completion of theseguidelines predated this approval (2). Foreach drug or drug class, the followingPopulation, Intervention, Comparator, andOutcomes (PICO) question was addressed:In patients with MDR-TB, are outcomessafely improved when regimens include thefollowing individual drugs or drug classescompared with regimens that do notinclude them?

The recommendations in this practiceguideline were supported by scientificevidence, including results of a propensityscore (PS)-matched individual patient datameta-analyses (IPDMA) conducted using adatabase of more than 12,000 patientrecords from 25 countries in support ofthese guidelines (see APPENDIX A:METHODOLOGY in the online supplement) (3).We used the Grading of Recommendations,Assessment, Development, and Evaluation(GRADE) approach to appraise the qualityof evidence and to formulate, write, andgrade most recommendations (4, 5).

AMERICAN THORACIC SOCIETY DOCUMENTS

e94 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Although published data on costs are notedin these guidelines, the committee did notconduct formal cost-effectiveness analyses ofthe treatments reviewed to determine arecommendation. The treatment of DR-TBcan be complicated and thus is necessarilypreceded and accompanied by importantcomponents of care relating to the access toTB experts, microbiological and moleculardiagnosis, education, monitoring and follow-up, and global patient-centered strategies.The writing committee considered that thesetopics are crucial but do not require formaland extensive evidence appraisal in thecontext of the present guidelines. FollowingGRADE guidance, it was thus decided thatthese practices would be addressed in goodpractice statements. Below, six ungradedgood practice statements as well as 25GRADE-based recommendations addressing21 PICO questions (Table 1) are listed.Questions were selected according to theirimportance to clinical practice, asdetermined by the guideline panel,expert advisors, and patient advocates.The implications of the strength ofrecommendation, conditional or strong, forpatients, clinicians, and policy makers aredescribed in APPENDIX A and shown inTable 2. Detailed and referencedinformation on treatment of MDR-TB isavailable below, including a summary of theevidence, and the benefits, harms, andadditional considerations of each practice orrecommendation.

Summary of Good PracticesFor patients being evaluated and treatedfor any form of drug-resistant TB, thefollowing six ungraded good practicestatements are emphasized, as the writingcommittee had high confidence in theirnet benefit:

1. Consultation should be requested with aTB expert when there is suspicion of orconfirmation of DR-TB. In the UnitedStates, TB experts can be found throughCDC-supported TB Centers ofExcellence for Training, Education, andMedical Consultation (http://www.cdc.gov/tb/education/rtmc/default.htm),through local health department TBcontrol programs (https://www.cdc.gov/tb/links/tboffices.htm), and throughinternational MDR-TB expert groupssuch as the Global TB Network (6).

2. Molecular DSTs should be obtained forrapid detection of mutations associated

with resistance. When rifampinresistance is detected, additional DSTshould be performed immediately forfirst-line drugs, fluoroquinolones, andaminoglycosides. Resistance to

fluoroquinolones should be excludedwhenever isoniazid resistance is found.

3. Regimens should include only drugs towhich the patient’s M. tuberculosisisolate has documented or high

Table 1. Questions Regarding the Treatment of Drug-Resistant Tuberculosis Selectedby the Guideline Writing Committee

Number of effective drugs in a regimen for MDR-TB1. Should patients with MDR-TB be prescribed five effective drugs vs. more or fewer

agents during the intensive and continuation phases of treatment?Duration of intensive and continuation phases of treatment for MDR-TB2. Should patients with MDR-TB undergoing intensive-phase treatment be treated for>6

mo after culture conversion or ,6 mo after culture conversion?3. Should patients with MDR-TB undergoing continuation-phase treatment be treated for

>18 mo after culture conversion or ,18 mo after culture conversion?Drug and drug classes for the treatment of MDR-TB4. In patients with MDR-TB, are outcomes safely improved when regimens include

amoxicillin/clavulanate compared with regimens that do not includeamoxicillin/clavulanate?

5. In patients with MDR-TB, are outcomes safely improved when regimens includebedaquiline compared with regimens that do not include bedaquiline?

6. In patients with MDR-TB, are outcomes safely improved when regimens includecarbapenems with clavulanic acid compared with regimens that do not include them?

7. In patients with MDR-TB, are outcomes safely improved when regimens includeclofazimine compared with regimens that do not include clofazimine?

8. In patients with MDR-TB, are outcomes safely improved when regimens includecycloserine compared with regimens that do not include cycloserine?

9. In patients with MDR-TB, are outcomes safely improved when regimens includedelamanid compared with regimens that do not include delamanid?

10. In patients with MDR-TB, are outcomes safely improved when regimens includeethambutol compared with regimens that do not include ethambutol?

11. In patients with MDR-TB, are outcomes safely improved when regimens includeethionamide/prothionamide compared with regimens that do not includeethionamide/prothionamide?

12. In patients with MDR-TB, are outcomes safely improved when regimens includefluoroquinolones compared with regimens that do not include fluoroquinolones?

13. In patients with MDR-TB, are outcomes safely improved when regimens include aninjectable compared with regimens that do not include an injectable?

14. In patients with MDR-TB, are outcomes safely improved when regimens includelinezolid compared with regimens that do not include linezolid?

15. In patients with MDR-TB, are outcomes safely improved when regimens includemacrolides compared with regimens that do not include macrolides?

16. In patients with MDR-TB, are outcomes safely improved when regimens includep-aminosalicylic acid compared with regimens that do not include p-aminosalicylic acid?

17. In patients with MDR-TB, are outcomes safely improved when regimens includepyrazinamide compared with regimens that do not include pyrazinamide?

Use of a standardized, shorter-course regimen of <12 mo for the treatment of MDR-TB18. In patients with MDR-TB, does treatment with a standardized MDR-TB regimen for

<12 mo lead to better outcomes than treatment with an MDR-TB regimen for 18–24 mo?Treatment of isoniazid-resistant, rifampin-susceptible TB:19a. Should patients with isoniazid-resistant TB be treated with a regimen composed of a

fluoroquinolone, rifampin, ethambutol, and pyrazinamide for 6 mo compared withrifampin, ethambutol, and pyrazinamide (without a fluoroquinolone) for 6 mo?

19b. Should patients with isoniazid-resistant TB be treated with a regimen composed offluoroquinolone, rifampin, and ethambutol for 6 mo and pyrazinamide for the first 2mo compared with a regimen composed of a fluoroquinolone, rifampin, ethambutol,and pyrazinamide for 6 mo?

Surgery as adjunctive therapy for MDR-TB:20. Among patients with MDR/XDR TB receiving antimicrobial therapy, does lung

resection surgery (i.e., lobectomy or pneumonectomy) lead to better outcomes thanno surgery?

Management of contacts exposed to an infectious patient with MDR-TB:21. Should contacts exposed to an infectious patient with MDR-TB be offered LTBI

treatment vs. followed with observation alone?

Definition of abbreviations: LTBI = latent TB infection; MDR=multidrug resistant; TB= tuberculosis;XDR= extensively drug resistant.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e95

likelihood of susceptibility (hereafterdefined as effective). Drugs known to beineffective based on in vitro growth-based or molecular resistance shouldNOT be used. This recommendationapplies to all drugs and treatmentregimens discussed in this practiceguideline, unless reliable methods oftesting susceptibility for a drug have yetto be developed.

4. Treatment response should bemonitored clinically, radiographically,and bacteriologically, with culturesobtained at least monthly for pulmonaryTB. When cultures remain positive after3 months of treatment, susceptibilitytests for drugs should be repeated.Weight and other measures of clinicalresponse should be recorded monthly.

5. Patients should be educated and askedabout adverse effects at each visit.Adverse effects should be investigatedand ameliorated.

6. Patient-centered case management helpspatients understand their diagnoses,understand and participate in theirtreatment, and discuss potentialbarriers to treatment. Patient-centeredstrategies and interventions shouldbe used to minimize barriers totreatment.

Summary of RecommendationsFor the selection of an effective MDR-TBtreatment regimen and duration of MDR-TB treatment:

1. We suggest using at least five drugs inthe intensive phase of treatment andfour drugs in the continuation phase oftreatment (conditional recommendation,very low certainty in the evidence).

2. We suggest an intensive-phase durationof treatment of between 5 and 7 monthsafter culture conversion (conditionalrecommendation, very low certainty inthe evidence).

3. We suggest a total treatment durationof between 15 and 21 months afterculture conversion (conditionalrecommendations, very low certainty inthe evidence).

4. In patients with pre–XDR-TB and XDR-TB, which are both subsets of MDR-TB,we suggest a total treatment duration ofbetween 15 and 24 months after cultureconversion (conditional recommendations,very low certainty in the evidence).

For the selection of oral drugs forMDR-TB treatment (in order of strength ofrecommendation):

5. We recommend including a later-generation fluoroquinolone(levofloxacin or moxifloxacin) (strongrecommendation, low certainty ofevidence).

6. We recommend including bedaquiline(strong recommendation, very lowcertainty in the evidence).

7. We suggest including linezolid(conditional recommendation, verylow certainty in the evidence).

8. We suggest including clofazimine(conditional recommendation, verylow certainty of evidence).

9. We suggest including cycloserine(conditional recommendation, verylow certainty in the evidence).

10. We suggest including ethambutol onlywhen other more effective drugscannot be assembled to achieve atotal of five drugs in the regimen(conditional recommendation, verylow certainty in the evidence).

11. We suggest including pyrazinamide ina regimen for treatment of patientswith MDR-TB or with isoniazid-resistant TB, when the M. tuberculosisisolate has not been found resistantto pyrazinamide (conditionalrecommendation, very low certainty inthe evidence).

12. The guideline panel was unable tomake a clinical recommendation for oragainst delamanid because of theabsence of data in the PS-matchedIPDMA conducted for this practiceguideline. We make a researchrecommendation for the conduct ofrandomized clinical trials and cohortstudies evaluating the efficacy, safety,and tolerability of delamanid incombination with other oral agents.Until additional data are available,the guideline panel concurs with theconditional recommendation of the2019 WHO Consolidated Guidelineson Drug-Resistant Tuberculosis

Table 2. Implications of Strong and Conditional Recommendations

Strong Recommendation (“We recommend . . .”)Conditional Recommendation

(“We suggest . . .”)

For patients The overwhelming majority of individuals in this situationwould want the recommended course of action, andonly a small minority would not.

The majority of individuals in this situation wouldwant the suggested course of action, but asizeable minority would not.

For clinicians The overwhelming majority of individuals should receivethe recommended course of action. Adherence to thisrecommendation according to the guideline could beused as a quality criterion or performance indicator.Formal decision aids are not likely to be needed to helpindividuals make decisions consistent with their valuesand preferences.

Different choices will be appropriate for differentpatients, and you must help each patient arrive at amanagement decision consistent with her or hisvalues and preferences. Decision aids may beuseful to help individuals make decisionsconsistent with their values and preferences.Clinicians should expect to spend more time withpatients when working toward a decision.

For policy makers The recommendation can be adapted as policy in mostsituations, including for the use as performanceindicators.

Policy making will require substantial debates andinvolvement of many stakeholders. Policies arealso more likely to vary between regions.Performance indicators would have to focus on thefact that adequate deliberation about themanagement options has taken place.

AMERICAN THORACIC SOCIETY DOCUMENTS

e96 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Treatment that delamanid may beincluded in the treatment of patientswith MDR/rifampin-resistant (RR)-TBaged >3 years on longer regimens (7).

For selected oral drugs previouslyincluded in regimens for the treatment ofMDR-TB:

13. We recommend NOT includingamoxicillin–clavulanate, with theexception of when the patient isreceiving a carbapenem wherein theinclusion of clavulanate is necessary(strong recommendation, very lowcertainty in the evidence).

14. We recommend NOT includingthe macrolides azithromycin andclarithromycin (strong recommendation,very low certainty in the evidence).

15. We suggest NOT includingethionamide/prothionamide if moreeffective drugs are available toconstruct a regimen with at leastfive effective drugs (conditionalrecommendation, very low certaintyin the evidence).

16. We suggest NOT includingp-aminosalicylic acid in a regimen ifmore effective drugs are available toconstruct a regimen with at leastfive effective drugs (conditionalrecommendation, very low certaintyin the evidence).

For the selection of drugs administeredthrough injection when needed to composean effective treatment regimen forMDR-TB:

17. We suggest including amikacin orstreptomycin when susceptibility tothese drugs is confirmed (conditionalrecommendation, very low certainty ofevidence).

18. We suggest including a carbapenem(always to be used with amoxicillin-clavulanic acid) (conditionalrecommendation, very low certainty ofevidence).

19. We suggest NOT including kanamycinor capreomycin (conditionalrecommendation, very low certainty inthe evidence).

A summary of the recommendationson drugs, the certainty in the evidence, andthe relative risks of success and death isprovided in Figure 1. Additional details andother outcomes of interest are provided inthe section on Drugs and Drug Classes and

RecommendationDrug / Drug

ClassFOR AGAINST

Certaintyin the

evidence

Relative(95% CI)

Death

Relative(95% CI)Success

Strong Very LowaOR 0.4

(0.3 to 0.5)aOR 2.0

(1.4 to 2.9)

Strong Very LowaOR 0.5

(0.4 to 0.6)aOR 3.8

(2.8 to 5.2)

Strong Very LowaOR 0.6

(0.5 to 0.7)aOR 4.2

(3.3 to 5.4)

Conditional Very LowaOR 0.3

(0.2 to 0.3)aOR 3.4

(2.6 to 4.5)

Conditional Very LowaOR 0.8

(0.6 to 1.0)aOR 1.5

(1.1 to 2.1)

Conditional Very LowaOR 0.6

(0.5 to 0.6)aOR 1.5

(1.4 to 1.7)

Conditional Very LowaOR 1.0

(0.8 to 1.2)aOR 2.0

(1.5 to 2.6)

Conditional Very LowaOR 0.8

(0.6 to 1.1)aOR 1.5

(1.1 to 2.1)

Conditional Very LowaOR 1.0

(0.9 to 1.2)aOR 0.9

(0.7 to 1.1)

Conditional Very LowaOR 0.7

(0.6 to 0.8)aOR 0.7

(0.5 to 0.9)

Conditional Very LowaOR 1.0

(0.5 to 1.7)aOR 4.0

(1.7 to 9.1)

Concur with WHO conditional recommendation

Conditional Very LowaOR 0.9

(0.8 to 1.0)aOR 0.8

(0.7 to 0.9)

Conditional Very LowaOR 1.1

(0.9 to 1.2)aOR 0.5

(0.4 to 0.6)

Conditional Very LowaOR 1.2

(1.1 to 1.4)aOR 0.8

(0.7 to 1.0)

Conditional Very LowaOR 1.4

(1.1 to 1.7)aOR 0.8

(0.6 to 1.1)

Strong Very LowaOR 1.6

(1.2 to 2.0)aOR 0.6

(0.5 to 0.8)

Bedaquiline

Fluoroquinolone:Moxifloxacin

Fluoroquinolone:Levofloxacin

Linezolid

Clofazimine

Cycloserine

Injectables:Amikacin

Injectables:Streptomycin

Ethambutol

Pyrazinamide

Delamanid

EthionamideProthionamide

Injectables:Kanamycin

P-AminosalicylicAcid

Injectables:Capreomycin

Macrolides:AzithromycinClarithromycin

Amoxicillin-clavulanate Strong Very Low

aOR 1.7(1.3 to 2.1)

aOR 0.6(0.5 to 0.8)

Injectables:Carbapenems w/clavulanic acid

Figure 1. Summary of recommendations on drugs for use in a treatment regimen for patients withmultidrug-resistant tuberculosis, including strength of recommendation, certainty in the evidence, andrelative effects on death and treatment success. Additional details and other outcomes of interest areprovided in the section on Drugs and Drug Classes, and in APPENDIX B: EVIDENCE PROFILES in the onlinesupplement. Success is defined as end of treatment cure or treatment completion. aOR=adjustedodds ratio; CI = confidence interval; WHO=World Health Organization.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e97

in APPENDIX B: EVIDENCE PROFILES in theonline supplement.

For the use of theWHO-recommendedstandardized shorter-course 9- to 12-monthregimen for MDR-TB:

20. The shorter-course regimen isstandardized with the use ofkanamycin (which the committeerecommends against using) andincludes drugs for which there isdocumented or high likelihood ofresistance (e.g., isoniazid, ethionamide,pyrazinamide). Although the STREAM(Standard Treatment Regimen of Anti-Tuberculosis Drugs for Patients withMDR-TB) Stage 1 randomized trialfound the shorter-course regimen to benoninferior to longer injectable-containing regimens with respect to theprimary efficacy outcome (8), theguideline committee cannot make arecommendation either for or againstthis standardized shorter-courseregimen, compared with longerindividualized all-oral regimens thatcan be composed in accordance withthe recommendations in this practiceguideline. We make a researchrecommendation for the conduct ofrandomized clinical trials evaluatingthe efficacy, safety, and tolerability ofmodified shorter-course regimens thatinclude newer oral agents, excludeinjectables, and include drugs forwhich susceptibility is documented orhighly likely.

For the role of surgery in the treatmentof MDR-TB:

21. We suggest elective partial lungresection (e.g., lobectomy or wedgeresection), rather than medical therapyalone, for adults with MDR-TBreceiving antimicrobial-based therapy(conditional recommendation, verylow certainty in the evidence). Thewriting committee believes this optionwould be beneficial for patients forwhom clinical judgement, supportedby bacteriological and radiographicdata, suggest a strong risk of treatmentfailure or relapse with medical therapyalone.

22. We suggest medical therapy alone,rather than including elective totallung resection (pneumonectomy), foradults with MDR-TB receivingantimicrobial therapy (conditional

recommendation, very low certainty inthe evidence).

For the treatment of isoniazid-resistantTB:

23. We suggest adding a later-generationfluoroquinolone to a 6-month regimenof daily rifampin, ethambutol, andpyrazinamide for patients withisoniazid-resistant TB (conditionalrecommendation, very low certainty inthe evidence).

24. In patients with isoniazid-resistant TBtreated with a daily regimen of a later-generation fluoroquinolone, rifampin,ethambutol, and pyrazinamide,we suggest that the duration ofpyrazinamide can be shortened to2 months in selected situations(i.e., noncavitary and lower burdendisease or toxicity from pyrazinamide)(conditional recommendation, verylow certainty in the evidence).

For the management of contacts topatients with MDR-TB:

25. We suggest offering treatment forlatent TB infection (LTBI) for contactsto patients with MDR-TB versusfollowing with observation alone(conditional recommendation,very low certainty in the evidence).We suggest 6 to 12 months oftreatment with a later-generationfluoroquinolone alone or with a seconddrug, on the basis of drug susceptibilityof the source-case M. tuberculosisisolate. On the basis of evidence ofincreased toxicity, adverse events, anddiscontinuations, pyrazinamide shouldnot be routinely used as the seconddrug.

In this guideline, we provide newrecommendations for treatment of MDR-TB and for treatment of isoniazid-resistantTB. On the basis of the evidence reviewconducted for this guideline, a clinicalstrategy tool for building a treatmentregimen for MDR-TB is provided.Additional detailed and referencedinformation on treatment of MDR-TB inspecial situations (children, pregnantwomen, and HIV-infected patients),approaches to treatment monitoring, the useof case management and directly observedtherapy, the role of therapeutic drugmonitoring, and key research priorities areprovided below.

Introduction

The American Thoracic Society (ATS), U.S.Centers for Disease Control and Prevention,European Respiratory Society (ERS), andInfectious Diseases Society of America(IDSA) have jointly developed these Drug-Resistant Tuberculosis guidelines. Thesenew recommendations are based on thecertainty in the evidence (also known as thequality of evidence) and developed based onthe evidence that was appraised usingGRADE methodology, which incorporatespatient values and costs as well as judgmentsabout trade-offs between benefits and harms(4, 5). A carefully selected panel of experts,screened for conflicts of interest, includingspecialists in pulmonary medicine,infectious diseases, pediatrics, primary care,public health, epidemiology, economics,pharmacokinetics, microbiology, systematicreview methodology, and patient advocacy,was assembled to assess the evidencesupporting each recommendation. Incontrast to prior analytic approaches,wherein systematic reviews of aggregatedata were used for decision-making, a newPS-matched meta-analysis of IPDMA from12,030 patients, in 50 studies, from 25countries with confirmed pulmonaryRR-TB was conducted for this guideline(3). Given the paucity of high-qualityrandomized controlled trials conducted inDR-TB, individual data from observationalstudies represent the next best level ofevidence for analyses. Nonetheless, thewriting committee noted that observationaldata are prone to bias and confounding.The 21 PICO questions and associatedrecommendations are summarized below,all appraised using GRADE methodology(see APPENDIX B). Questions were selectedaccording to their importance to clinicalpractice, as determined by the guidelinepanel, expert advisors, and patientadvocates. The implications of thestrength of recommendation, conditional orstrong, for patients, clinicians, and policymakers are described in APPENDIX A. On thebasis of the GRADE methodologyframework, all recommendations in theseguidelines are based on very low certaintyin the evidence. The writing committeeselected death, treatment success, andserious adverse effects as the endpoints ofcritical importance on which to generaterecommendations. Our meta-analyticapproaches included PS matching to

AMERICAN THORACIC SOCIETY DOCUMENTS

e98 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

reduce confounding (on the basis ofindividual-level covariates of age, sex, HIVcoinfection, acid-fast bacilli [AFB] smearresults, cavitation on chest radiographs,history of TB treatment with first-line orsecond-line TB drugs, and number ofpossibly effective drugs in the regimen,among other variables), described in detail inthe publication of the PS-matched IPDMApublication and in APPENDIX A (3). However,the risk of bias remained serious, because theaverage loss to follow-up across includedstudies was 10% to 20%. In addition, despitethe efforts, there was a large residualimbalance in background regimens used inexperimental and control groups.

These guidelines are intended forsettings in which treatment is individualizedand where mycobacterial cultures, molecular(genotypic) and culture-based (phenotypic)DSTs, and radiographic facilities are available(9, 10). Of note, published data on costs arereferenced in these guidelines, but thecommittee did not conduct formal cost-effectiveness analyses of the treatments andinterventions reviewed in relation todetermining a recommendation.

In these guidelines, MDR-TB is definedspecifically as resistance to at least isoniazidand rifampin, the two most importantfirst-line drugs. XDR-TB is a subset ofMDR-TB with additional resistance to afluoroquinolone and a second-line injectableagent. Because XDR-TB evolves fromMDR-TB in two steps, the term “pre–XDR-TB” was introduced to identify MDR-TBwith additional resistance to either one butnot both of these classes of drugs. In theseguidelines, we also provide recommendationsfor the treatment of isoniazid-resistant TB.

Good Practices for TreatingDR-TB

The fundamentals of TB care, regardless ofthe treatment selected, rest on ensuringtimely diagnosis and initiation ofappropriate therapy, with ongoing supportand management to achieve successfultreatment completion and cure. Theresponsibility for successful treatment of TBis placed primarily on the provider orprogram initiating therapy rather than onthe patient (11). Nevertheless, a patient-centered approach, described more fully inthe section on case management below,requires the involvement of the patient indecision-making. We recommend seeking

consultation with an expert in TB whenthere is suspicion for or confirmationof DR-TB (ungraded good practicestatement). In the United States, DR-TBexperts can be found through CDC-supported TB Centers of Excellence forTraining, Education, and MedicalConsultation (http://www.cdc.gov/tb/education/rtmc/default.htm), throughlocal health department TB ControlPrograms (https://www.cdc.gov/tb/links/tboffices.htm), and throughinternational MDR-TB expert groups suchas the British Thoracic Society MDR-TBClinical Advisory Service (http://mdrtb.brit-thoracic.org.uk/) and the Global TBNetwork (6). Additional good practices inthe treatment of patients in need ofevaluation for DR-TB include the following:

Diagnosing TB and Identification ofDrug ResistanceThe potential for drug resistance isconsidered in every patient. An aggressiveeffort is made to collect biological specimensfor detection ofM. tuberculosis and for drugresistance. A rapid test for a least rifampinresistance should ideally be done for everypatient, but especially for those at risk ofdrug resistance. The concern for possibleresistance is heightened for patients fromareas of the world with at least a moderateincidence of TB in general (>20/100,000)and a high primary MDR-TB prevalence(>2%) (12). Individuals who have orrecently had close contact with a patientwith infectious DR-TB, especially when thecontact is a young child or has HIVinfection, are at risk of developing DR-TB.Molecular methods, and more recentlywhole-genome sequencing (WGS), areincreasingly available and can provideinformation on resistance to all first-lineand many second-line drugs. Many publichealth laboratories provide moleculartests, and WGS is available in selectedlaboratories. These tests can be used toguide initial therapeutic decisions andcontribute to population-level control ofDR-TB. Providers should be familiar withphenotypic and genotypic laboratoryservices available in their locale. In theUnited States, CDC’s Division ofTuberculosis Elimination LaboratoryBranch provides testing services forboth clinical specimens and isolates ofM. tuberculosis (https://www.cdc.gov/tb/topic/laboratory/default.htm).CDC’s Molecular Detection of Drug

Resistance (MDDR) service serves torapidly identify DR-TB. This service usesDNA sequencing for detection of mutationsmost frequently associated with resistanceto both first-line (e.g., rifampin, isoniazid,ethambutol, and pyrazinamide) as wellas second-line drugs (MDDR ServiceRequest Form is available here: https://www.cdc.gov/tb/topic/laboratory/MDDRsubmissionform.pdf). Recentlypublished ATS/CDC/IDSA Official PracticeGuidelines for the diagnosis of TB provideadditional details on the optimal use ofdiagnostic tools and algorithms (13).

Treatment and Monitoring of DR-TBRegimens should only include drugs towhich the patient’s isolate has documentedor high likelihood of susceptibility. Drugsknown to be ineffective, on the basis ofin vitro resistance or clinical andepidemiological information (i.e., resistancein the index case or high populationprevalence of resistance), should not beused, even when resistance is present inonly a small percentage of the mycobacteriain the population. If at least 1% oforganisms in a solid media culture exhibitresistance to a drug (the current standardlaboratory definition of drug resistance)(14), using that drug in a regimen willincrease the risk for poor treatmentoutcomes, and the isolate will eventuallyexhibit 100% resistance to the drug. Drugsshould be selected based on their efficacyand the likelihood that patients will be ableto tolerate them without significant toxicity.Treatment response is monitored clinically(decrease in cough and systemic symptomsand increase in weight), radiographically,and bacteriologically (15–18). If sputumcultures remain positive after 3 months oftreatment, or if there is bacteriologicalreversion from negative to positive at anytime, DST should be repeated (11). Patientsshould be asked about the clinical responseat each visit and weight recorded monthly.Monthly cultures help to identify earlyevidence of failure (19). Most persons havedifficulty taking one or more of the drugsused to treat MDR-TB. Patients should beeducated about adverse effects, and alladverse effects should be investigated andameliorated. Some adverse effects aredifficult to tolerate but do not put patientsat risk for serious short- or long-termdamage to organ systems and can bemanaged with symptom-specific ancillarymedication and supportive care. Nausea

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e99

with vomiting is common and is not alwaysan indication to discontinue therapypermanently. New-onset vomiting mayindicate liver toxicity or, in childrenespecially, increased intracranial pressure. Ifdrug-induced liver toxicity (and increasedintracranial pressure) is excluded, vomitingcan be managed by changing dosingschedule, giving medications with a smallsnack (noting that this may affectplasma concentrations of the drug), orpremedicating adult patients with anantiemetic (noting that some prolong theQT interval) before the dose. Patients maynote fatigue or describe myalgia orarthralgia, but these symptoms are nottypically treatment limiting. Althoughlow-grade adverse effects can resolvegradually over time, recognizing thenegative impact on patient’s quality of life,all adverse effects must be addresseddiligently. The Curry International TBCenter’s guides, Drug Resistant TB:Clinician’s Survival Guide and the NursingGuide for Managing Side Effects to Drug-Resistant TB Treatment, in addition to theWorld Health Organization (WHO)companion handbook, are good resourcesto assist in evaluation and management ofpatients (15, 16, 18).

Infection Control and DR-TBThree main strategies will reduce thetransmission of DR-TB: rapid diagnosis,prompt appropriate treatment, andimproved airborne infection control(11, 13, 20, 21). Rapid molecular DST andconventional phenotypic culture-basedDST are almost universally available in theUnited States, Europe, and low-incidence,high-resource countries (22). Targetedactive case finding combined with rapiddiagnostics leading to effective therapy is astrategy endorsed by WHO (13, 23, 24).

Treatment delays have been associatedwith increased transmission. A systematicreview and meta-analysis looking at patient-related risk factors for transmission of M.tuberculosis found that treatment initiationdelays of 28 to 30 days were significantlyassociated with increased transmission tocontacts (25). Effective therapy renderspatients with TB, even those with DR-TB,rapidly noninfectious (24, 26). The rapidreduction in infectiousness even in thesetting of MDR-TB makes outpatienttherapy possible, but directly observedtherapy (DOT) applied through patient-centered approaches plays an important

role in this regard (11, 23). Infectioncontrol measures such as administrativeand environmental controls, and personalprotective equipment, listed in order ofpriority, are important for preventing thetransmission of M. tuberculosis regardlessof drug susceptibility. Every healthcarefacility should have these measures in placeper CDC guidelines (27). Furthermore,patients should be educated about theimportance of infection control measures,such as the value of good ventilation, openwindows, and the risks of exposure forchildren ,5 years of age andimmunocompromised individuals (21).

Case Management for DR-TBCase management is a collaborative processthat entails engaging with patients;comprehensively assessing, monitoring,and attending to patients’ physical,psychological, social, material, andinformational needs; care planning;medication management; facilitating accessto services; and functioning as patientadvocate/agent (28–30). Commonly, casemanagement is used in community andpublic health settings to coordinate servicesfor patients with chronic and complexhealth conditions and to attain good,quality, cost-effective outcomes. Thepractice of case management has long beenconsidered an important component ofcare for patients with TB (20, 31, 32), and apatient-centered (or family-centered in caseof children) approach is preferred (7, 11,33). Patient-centered case managementhelps patients understand theirdiagnosis and treatment and participatein treatment selection and promotescommunication about factors that matterto the patient (33–35). Importantly, apatient-centered approach includesdiscussions with the patient to identifypotential barriers to care and the selectionof strategies and interventions to addressand minimize these barriers (7, 11, 33,35–37).

Case management tools for drug-resistant TB include a drug-o-gram, whichorganizes clinical details in a format thataligns test results with treatment, as well aslaboratory, bacteriology, and other toxicitymonitoring flow sheets. This format is avisible representation of pertinent clinicalparameters being tracked (16, 38). Amonitoring checklist or care plan can alsohelp the case manager ensure timely drugtoxicity monitoring and provision of

examinations required to assess a patient’sresponse to treatment. Case managementinterventions that have demonstratedfavorable outcomes include the provision ofpatient education and counseling related todiagnosis, treatment, and adherence, as wellas the use of treatment adherenceinterventions alongside suitable patient-centered administration options. Forexample, home or community-based DOTis shown to be preferred and associatedwith greater likelihood of treatment successcompared with health facility–based DOTor self-administered therapy (11, 39).Enhancing treatment completion throughthe use of patient-centered casemanagement strategies, including DOT,also aims to reduce risk of acquisition ofdrug resistance, which aligns withinternational efforts in antibioticstewardship (1). Recent WHO guidelinesevaluated various adherence interventionsand outcomes of TB treatment through asystematic review of clinical trials andobservational studies, identifying data from129 studies for quantitative analysispublished through 2018 (39). Anothermeta-analysis of 22 randomized controlledtrials of DOT and other interventions toimprove adherence reported significantincreases in cure with DOT (18%) and withpatient education and counseling (16%).Compared with the complementary groups,loss to follow-up was 49% lower with DOT,26% lower with financial incentives, and13% lower with patient education andcounseling. However, there was nosignificant reduction in mortality (40).

The use of video-enabled electronicdevices to conduct DOT is expandingrapidly (41, 42). Electronic methods forDOT have the potential to improve TBtreatment outcomes and extend publichealth support to patients with TB whenface-to-face DOT is not feasible. Pilotstudies have reported that patients findelectronic methods for DOT to be bothacceptable and more convenient thantraditional in-person DOT. These studiesalso reported good adherence, fewerunobserved doses, and high satisfactionamong study participants (43–46). Furtherevidence is needed to validate electronicallyobserved therapy under more diverseconditions, with larger cross-sections ofpatient subgroups, including children, andto determine to what degree these methodsimpact treatment outcomes for patientswith DR-TB (39, 47–49).

AMERICAN THORACIC SOCIETY DOCUMENTS

e100 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Treatment of MDR-TB,Number of Drugs, andDuration of TreatmentPhases

Number of Drugs in the Regimen

Up until the recent 2019 WHOConsolidated Guidelines on Drug-ResistantTuberculosis Treatment (7), WHO hasrecommended at least five drugs in theintensive phase of treatment, defined by theuse of a second-line injectable agent (23).The recent WHO change to recommendingat least four effective drugs at initiation of

treatment is graded as a conditionalrecommendation with very low certainty inthe estimates of effect (7). Of note, both ourguideline committee and the 2019 WHOguidelines promote the use of newer orrepurposed oral agents with greater efficacyand deemphasize the use of injectableagents (7). Given these changes and that aninjectable drug is no longer obligatory, theintensive phase can no longer be definedby the inclusion of injectables. In thisguideline, we define the intensive phaseto be the initiation phase of treatment withat least five effective drugs. Theserecommendations do not apply to theWHO-endorsed shorter-course 9- to12-month (Bangladesh) regimen, whichcombines seven drugs for 4 months or untilsputum smear conversion, whichever is thelonger period, followed by four drugs in thecontinuation phase (23).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B. Forthe analysis of number of effective drugs,we counted drugs with published evidencefrom randomized trials of effectiveness (3).Hence, we counted ethambutol,pyrazinamide, all injectables andfluoroquinolones, ethionamide/prothionamide,cycloserine/terizidone, and p-aminosalicylicacid on the basis of DST showing

susceptibility and counted clofazimine,linezolid, carbapenems, bedaquiline,and delamanid—if susceptible, or noDST for that drug. We did not countamoxicillin/clavulanate (in the absence of acarbapenem) and macrolides as effectivedrugs (see PICO Questions 4 and 15).The intensive phase was defined by theuse of an injectable agent (other thanimipenem–cilastatin or meropenem).Culture conversion was not an outcomemeasure. Instead, the analysis assessed theassociation of the number of possiblyeffective drugs included in the first 2 weeksof the intensive phase with the two finaltreatment outcomes: 1) “treatment success,”which includes both cure and treatmentcompleted; and 2) mortality. For theintensive phase of treatment, final treatmentoutcomes were compared between thosetreated with five or more (n= 2,527) effectivedrugs and those treated with three or foureffective drugs (n= 5,923).

Benefits. In our PS-matched IPDMA,using zero to two effective drugs as thereference value, treatment success wasmost likely with regimens for MDR-TBcontaining five effective drugs in theintensive phase (Table 3). Mortality wasalso significantly reduced for those takingfive or six effective drugs.

The number of drugs used in thecontinuation phase was also evaluated(Table 4). In the continuation phase, four

Table 3. Propensity Score–matched Analysis of the Number of Drugs in the Intensive Phase of Treatment and the aOR of TreatmentSuccess versus Failure or Relapse

No. of Drugs

No. of Patients Propensity Score–matched Analysis

Success/Total (%) Death/Total (%) aOR (95% CI)Risk Difference(95% CI) (%)

For the analysis of success vs. fail*/relapse†

0–2 drugs 1,097/1,236 (88.8) — 1.0 (reference)3 drugs 1,257/1,407 (89.3) — 1.7 (1.4 to 2.0) 6 (4 to 7)4 drugs 1,657/1,847 (89.7) — 1.2 (1.4 to 2.0) 8 (6 to 9)5 drugs 926/986 (93.9) — 3.0 (2.3 to 3.9) 8 (7 to 10)>6 drugs 523/568 (92.1) — 2.3 (1.6 to 3.1) 4 (1 to 7)

For the analysis of death vs. success/fail*/relapse†

0–2 drugs — 205/1,441 (14.2) 1.0 (reference)3 drugs — 233/1,640 (14.2) 0.9 (0.8 to 1.1) 21 (23 to 1)4 drugs — 345/2,192 (15.7) 1.1 (0.9 to 1.2) 23 (25 to 21)5 drugs — 104/1,090 (9.5) 0.6 (0.5 to 0.7) 22 (23 to 0)>6 drugs — 54/622 (8.6) 0.5 (0.4 to 0.7) 2 (20 to 5)

Definition of abbreviations: aOR=adjusted odds ratio; CI = confidence interval.*World Health Organization definitions: fail = treatment terminated or need for permanent regimen change of at least two antituberculosis drugs becauseof: lack of conversion by the end of the intensive phase, bacteriological reversion in the continuation phase after conversion to negative, evidence ofadditional acquired resistance to fluoroquinolones or second-line injectable drugs or adverse drug reactions.†Relapse was defined as a positive bacteriological culture in the 12 months after treatment completion.

PICO Question 1: Should patients withMDR-TB be prescribed five effectivedrugs versus more or fewer agentsduring the intensive and continuationphases of treatment?

Recommendation 1a: We suggestusing at least five drugs in theintensive phase of treatment of MDR-TB (conditional recommendation,very low certainty of evidence).

Recommendation 1b: We suggestusing at least four drugs in thecontinuation phase of treatmentof MDR-TB (conditionalrecommendation, very low certainty ofevidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e101

drugs gave the greatest success (adjustedodds ratio [aOR], 2.3) and four or more drugsthe greatest reduction in mortality.

Few significant differences were foundin patients with pre–XDR-TB or XDR-TB.Subgroup analyses in pre–XDR-TB andXDR-TB did not suggest that a differentnumber of effective drugs for the intensiveand continuation phases was required toachieve good outcomes.

Harms. The recommendation places ahigher value on reduced morbidity andgreater treatment success than on the as yetunmeasured avoidance of the summative orsynergistic toxicity across drugs in thetreatment regimen.

Additional considerations. Each drugwas counted as equivalent in effectiveness,representing the average effectiveness of themost widely used MDR-TB drugs, for thepurpose of adjusting for the number of othereffective drugs in the regimen. A regimencomposition approach with suggestedranking of drugs is provided in the sectionBUILDING A TREATMENT REGIMEN FOR MDR-TB. The smallest number of drugs toachieve sputum conversion and ensure acure without relapse minimizes problemswith drug–drug interactions, reducesadverse effects, and is cost effective.Injectable drugs are problematic in terms ofroute of administration, patient preference,significant adverse effects such as hearingloss or vertigo, and costs of administration

and monitoring blood levels. On the basisof these limitations, combined withevidence of low efficacy for injectableagents, the use of injectables in defining theintensive phase of treatment is no longerappropriate. WHO has recategorizedinjectables to a lower grouping andrecommended that the injectables only beused when oral medicines from highercategories cannot be used (7). The SouthAfrican National Department of Health alsohas endorsed replacing the injectable agentsin the shorter-course MDR-TB treatmentregimen in adults and children .12 yearsof age with bedaquiline, on the basis of aretrospective observational study ofimproved mortality with adding bedaquiline(50). In conjunction, experts from theSentinel Project on Pediatric Drug-ResistantTuberculosis (http://sentinel-project.org)have also called for replacing the injectabledrug in children ,12 years of age with analternative effective drug (51).

Conclusions. At least five drugs should beused in the intensive phase of treatment and fourdrugs in the continuation phase of treatment ofMDR-TB (conditional recommendation, verylow certainty of evidence). Drugs of poor ordoubtful efficacy should not be added to aregimen purely to ensure that the recommendednumber of drugs is obtained.

Research needs. Randomizedcontrolled trials with fewer but moreeffective and safer drugs should be

undertaken (e.g., TB-PRACTECAL,[Pragmatic Clinical Trial for a MoreEffective Concise and Less Toxic MDR-TBTreatment Regimen] ClinicalTrials.govidentifier NCT02589782). Randomizedtrials comparing regimens with andwithout injectable agents are underway(e.g., STREAM-stage 2, NCT02409290;Evaluating Newly Approved Drugsfor Multidrug-resistant TB: endTB,NCT02754765). The effect of duration ofindividual drugs in the intensive phase onculture conversion as well as treatment curewithout relapse should be explored (52).

Duration of Intensive andContinuation Phases in TreatingMDR-TBTreatment of both drug-susceptible andMDR-TB has typically been divided across

PICO Question 2: Should patients withMDR-TB undergoing intensive-phasetreatment be treated for >6 monthsafter culture conversion or ,6 monthsafter culture conversion?

Recommendation 2: In patients withMDR-TB, we suggest an intensive-phase duration of treatment ofbetween 5 and 7 months afterculture conversion (conditionalrecommendation, very low certainty ofevidence).

Table 4. Propensity Score–matched Analysis of the Number of Drugs in the Continuation Phase of Treatment and the aOR ofTreatment Success versus Failure or Relapse

No. of Drugs

No. of Patients Propensity Score–matched Analysis

Success/Total (%) Death/Total (%) aOR (95% CI)Risk Difference(95% CI) (%)

For the analysis of success vs. fail*/relapse†

0–1 drug 1,017/1,144 (88.9) — 1.0 (reference)2 drugs 1,272/1,425 (89.2) — 1.1 (0.9 to 1.3) 1 (21 to 3)3 drugs 1,623/1,810 (89.7) — 1.2 (1.0 to 1.4) 3 (1 to 5)4 drugs 816/864 (94.4) — 2.3 (1.7 to 3.1) 3 (1 to 5)>5 drugs 346/383 (90.3) — 1.2 (0.9 to 1.8) 24 (28 to 21)

For the analysis of death vs. success/fail*/relapse†

0–1 drug — 187/1,331 (14.0) 1.0 (reference)2 drugs — 193/1,618 (11.9) 0.8 (0.6 to 0.9) 23 (25 to 21)3 drugs — 307/2,117 (14.5) 1.0 (0.8 to 1.1) 24 (25 to 22)4 drugs — 78/942 (8.3) 0.5 (0.4 to 0.7) 21 (24 to 1)>5 drugs — 37/420 (8.8) 0.5 (0.4 to 0.8) 5 (1 to 8)

Definition of abbreviations: aOR=adjusted odds ratio; CI = confidence interval.*World Health Organization definitions: fail = treatment terminated or need for permanent regimen change of at least two anti-tuberculosis drugs becauseof: lack of conversion by the end of the intensive phase, bacteriological reversion in the continuation phase after conversion to negative, evidence ofadditional acquired resistance to fluoroquinolones or second-line injectable drugs or adverse drug reactions.†Relapse was reported as a positive bacteriological culture in the 12 months after treatment completion.

AMERICAN THORACIC SOCIETY DOCUMENTS

e102 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

two phases, with an initial “intensive” phasethat contains more drugs than thesubsequent “continuation” phase. ForMDR-TB, this intensive phase hashistorically been characterized by theuse of an aminoglycoside (amikacinor kanamycin) or a polypeptide(capreomycin) delivered parenterally (53).This approach has been intended to providegreater bactericidal activity during the timewhen the bacillary burden is highest andreducing the number of drugs in thecontinuation phase to reduce risk oftoxicity and intolerability caused by themultidrug regimen at a phase whenthe microbial burden has diminished. The6-month duration of bedaquiline treatment,in part, was developed by analogy, andclinical trials are underway to determinewhether this drug may replaceaminoglycosides in terms of an initialintensive phase of treatment. The durationof the intensive phase has never beenexamined through a randomized,controlled clinical trial. Rather, it has beendefined by a combination of practicality,the clinical experiences of MDR-TBexperts, and, most recently, an IPDMA thatresulted in the WHO practice guidelinerecommendations of 2011 that an intensivephase of at least 8 months’ duration be used(54). This represented a departure from theprior recommendations in 2008 that theinjectable agent should be continued for atleast 6 months and at least 4 months afterthe patient first becomes and remainssmear or culture negative (55). This issimilar to current expert guidance providedby the Curry International TB Center DrugResistant TB: Clinician’s Survival Guide,namely “Intensive phase: recommend atleast 6 months beyond culture conversionfor the use of injectable agent” (16). ThePS-matched IPDMA completed as part ofthe present guideline development processincludes cohorts that were treatedaccording to this range of durations of theintensive phase and represents the available

evidence base for our recommendation. Ouranalyses and recommendations for theduration of intensive and continuation phasesof therapy are anchored to the timing ofculture conversion, as this approach factors inthat treatment response may vary by patient,resistance patterns, and regimen compositionand potency, among other factors.

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B.The association between treatmentduration of the intensive phase (defined bythe use of an injectable agent) after cultureconversion and outcomes (success, failure,relapse) was examined among 4,122(49.3%) subjects from 29 studies whohad sufficient information to be included inthis analysis (3). This total reflectedexclusion of studies without a reportedintensive-phase duration (n= 942 [11.3%]patients) or time to culture conversion(n= 1,880 [22.5%] patients), as well as 955(11.4%) patients in the included studies butfor whom intensive-phase duration or timeto conversion was not known. Patientstaking the 9- to 12-month standardizedshorter-course regimen were excluded (seePICO Question 18). An additional 464(5.0%) patients were excluded for conversionafter the end of the intensive phase or after14.3 months, or for an intensive phaseexceeding 24.3 months. Among the 4,122patients included, 3,303 (80.1%) had MDR-TB without resistance to fluoroquinolone orsecond-line injectable. The remainder hadpre- or XDR-TB. To reflect meaningfulvariations in the duration of the intensiveregimen after conversion, stratified analyseswere conducted: strata were 0 to 1.0 months,1.01 to 3.0 months, 3.01 to 5.0 months, 5.01to 7.0 months, and 7.1 to 15.0 months (seeTable 5). The largest proportion (28.6%[1,179]) received treatment for 5.01 to 7(mean, 5.9) months after conversion. Timeto conversion was roughly inversely

proportional to duration of intensivephase, suggesting that the tendency inthe included studies was to treat for atotal intensive-phase duration of 5 to8 months.

Benefits. The patients who receivedtreatment for 5.01 to 7 months afterconversion experienced a 3.3-fold increasein adjusted odds of treatment success (95%confidence interval [CI], 2.1–5.2) comparedwith the reference group (0–1.0 months ofintensive phase after conversion). Somevariability was observed in distribution ofintensive-phase duration after cultureconversion by baseline characteristics:,10.0% of patients who received longerdurations of intensive phase were HIV-coinfected, compared with .15% in the 0.0to 3.0 month interval groups. Number of(effective) drugs was greater in longerdurations than in shorter.

Patients with MDR-TB who received5.01 to 7.0 months of intensive-phasetreatment after conversion had the highestodds of success in adjusted PS-matchedanalysis; although other, shorter intervalsshowed improvement over the reference,<1.0 month of postconversion treatment,the benefits of 5 to 7 months of treatmentafter conversion were more pronounced(Table 6). The effect estimates werealso better for subgroups of MDRonly (aOR, 2.0; 95% CI, 1.1–3.4) andpre-XDR (aOR, 1.5; 95% CI, 0.6–3.7)subgroups.

HarmsDetailed data on duration-related toxicitieswere not available through our PS-matchedIPDMA. On the basis of the clinicalobservations and experiences of the MDR-TB experts on the guideline committee, theintensive phase should not be prolongedbeyond that considered necessary tooptimize treatment outcomes.

Additional considerations. Noinformation is available in the datasets on

Table 5. Distribution of 4,122 Patients by Duration of Intensive Phase after Culture Conversion

Interval from Culture Conversion toEnd of Intensive Phase (Mean) (mo)

No. ofPatients (%)

Mean Time toConversion (mo)

Mean Total IntensivePhase Duration (mo)

Mean Total TreatmentDuration (mo)

0–1.0 (0.6) 251 (6.1) 4.7 5.3 20.01.01–3.0 (2.3) 695 (16.9) 3.0 5.3 20.13.01–5.0 (4.2) 917 (22.2) 2.1 6.4 20.05.01–7.0 (5.9) 1,179 (28.6) 1.7 7.6 20.67.01–15.0 (10.5) 1,080 (26.2) 2.0 12.5 22.9

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e103

how duration was selected. Durationselected may reflect interim response totreatment or may reflect a planned durationthat is not conditioned on treatmentresponse; the latter is suggested by theobservation that duration was inverselyrelated to time to conversion except in the7.01- to 15-month interval. Analyses wereadjusted for possible baseline confounders,relying on PS matching, to reduce the biasintroduced. However, the possibility ofunmeasured confounding by indication, inparticular by time-varying characteristics(such as toxicity, microbiological,radiographic, or clinical results), cannot beruled out. Such confounding by indicationwould likely result in an underestimate of thebenefit of a longer intensive phase afterconversion: patients in the 7.01- to 15-monthinterval had slower conversion and morepoor outcomes than those in the 5.01- to 7.0-month interval. Last, the optimal totalduration of treatment for MDR-TB usinginjectable-free, all-oral regimens cannot bedetermined from these datasets, but clinicaltrials evaluating newer drugs and all-oralregimens for MDR-TB are underway (56).

Conclusions. We suggest an intensive-phase treatment of between 5 and 7 monthsafter culture conversion in patients withMDR-TB (conditional recommendation,very low certainty in the evidence). Theclinical context, extent of disease, andresponse to treatment, among other factors,will play a role in choosing a final durationfrom within the recommended range. Therewere limited data in pre–XDR-TB and XDR-TB. Subgroup analyses in pre-XDR andXDR-TB did not suggest that a differentduration for the intensive phase would berequired to achieve good outcomes. Thisintensive-phase duration recommendationdoes not apply to the 9- to 12-month shorter-course regimen addressed in PICO 18.

Research needs. Further research isurgently needed to define the optimaldurations of treatment using newer drugsand all-oral regimens. Research in definingoptimal duration that prevents death or lossto follow-up is also needed. In the shortterm, further research on datasets thatinclude longitudinal observations thatcan inform choice of regimen durationwould be important to reducing theuncertainty around the presentrecommendations. In the long term,randomized controlled trials evaluatingvarious intensive phases and durations areneeded. Research on stratified medicineapproaches that use measures of burden ofdisease and consider subgroups in theselection of the optimal duration may allowfor greater precision and better informdecision-making around the balance ofbenefits and harms of durations forindividual patients (57).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B. Theassociation between total duration of treatmentafter culture conversion and outcomes(success, failure, relapse) was performed on4,691 observations from 32 studies (3). Weexcluded 18 studies (n=2,615) that did not

report time to culture conversion and 798patients from included studies becauseendpoints were missing. Last, 259 patientswere excluded for outlier values in totalduration or culture conversion; patients whoseoutcome was loss to follow-up or death wereexcluded. Among the 4,691 included, 3,034had MDR-TB. The remainder had pre-XDRor XDR-TB. To reflect meaningful variationsin the duration of treatment after conversion,stratified analyses were conducted: Time toconversion was inversely related to duration oftreatment after conversion (Table 7).

Benefits. Among all resistance groups,durations of 15.01 to 21months after conversionoutperformed the reference (12.01–15 mo);intervals of 15.01 to 18 (aOR, 2.1; 95% CI,1.4–3.1), 18.01 to 21 (aOR, 1.6; 95% CI,1.1–2.3), and 21.01 to 24 (aOR, 1.2; 95% CI,0.9–1.8) months were indistinguishable fromeach other (Table 8). In the MDR-onlysubgroup, the duration of 15.01 to 18.0 monthswas associated with similar to slightly improvedoutcomes compared with reference (aOR, 1.8;95% CI, 1.0–3.0) (data not shown). Results forthe pre–XDR-TB subgroup supported thenotion that longer intervals were associatedwith success; effects were statistically significant,with confidence intervals that were wide andoverlapping across durations of 15.01 to 24months (data not shown).

Table 6. Adjusted Estimates of Treatment Success by Duration of Intensive Phase after Culture Conversion, All Forms ofMultidrug-Resistant Tuberculosis (N=4,122)

Intervals from Sputum CultureConversion to End ofIntensive-Phase Treatment (mo)

No. of Patients Propensity Score–matched Analysis

TreatmentSuccess Total No. of Pairs aOR 95% CI

Risk Difference(95% CI)

0–1.0 239 251 — 1.0 Reference —1.01–3.0 668 695 694 1.5 1.0 to 2.3 0.02 (0.00 to 0.03)3.01–5.0 878 917 906 1.4 1.0 to 2.0 0.02 (0.00 to 0.03)5.01–7.0 1,158 1,179 1,179 3.3 2.1 to 5.2 0.04 (0.03 to 0.05)7.01–15.0 1,025 1,080 1,079 1.1 0.8 to 1.5 0.01 (20.01 to 0.02)

Definition of abbreviations: aOR=adjusted odds ratio; CI = confidence interval.

PICO Question 3: Should patients with MDR-TB undergoing continuation-phasetreatment be treated for >18 months after culture conversion or ,18 months afterculture conversion?

Recommendation 3a: In patients with MDR-TB, we suggest a total treatmentduration of between 15 and 21 months after culture conversion (conditionalrecommendations, very low certainty in the evidence).

Recommendation 3b: In patients with pre–XDR-TB and XDR-TB, we suggest a totaltreatment duration of between 15 and 24 months after culture conversion(conditional recommendations, very low certainty in the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e104 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Harms. Extending treatment longerthan necessary can engender additionaltoxicity and costs to patients and healthsystems. For this reason, recommendationsare for the minimum duration found to havea significant treatment advantage, anddifferent recommendations are made for theMDR only and XDR/pre-XDR subgroups.

Additional considerations. Noinformation on how duration was selected isavailable from the data. Duration selectionmay reflect interim response to treatment ormay reflect a planned duration that is notconditioned on treatment response; the latteris suggested by the observation that durationwas inversely related to time to conversionexcept in the longest interval. Analyses wereadjusted for all possible baselineconfounders, relying on PS matching, tominimize the bias introduced. Nevertheless,the possibility of unmeasured confounding byindication, in particular by time-varyingcharacteristics, cannot be ruled out. Suchconfounding by indication would likelyresult in an underestimate of the benefit of alonger treatment duration after conversion.No significant difference in outcomes wasobserved across durations for the XDR-TBsubgroup, likely because of small numbersand the aforementioned potential bias.

Conclusions. We suggest a totaltreatment duration of at least 15 and up to21 months after culture conversion for patientswith MDR-TB only. In patients with pre–XDR-TB and XDR-TB, we suggest a total treatmentduration of between 15 and 24 monthsafter culture conversion (conditionalrecommendations, very low certainty in theevidence). The clinical context and extent ofdisease will be relevant for choosing a finalduration from within the recommended range.

Research needs. The optimal totalduration of regimens composed of onlyoral drugs urgently needs further research.Defining optimal total durations thatprevent death or loss to follow-up alsoneeds study. In the short term, furtherresearch on datasets that includelongitudinal observations that can informchoice of regimen duration would be veryimportant to reducing the uncertaintyaround the present recommendation. Inthe long term, randomized controlledtrials comparing different intensive phasesand durations are needed. As with theduration of the intensive phase, research isneeded in stratified medicine approachesthat consider measures of burden ofdisease, and subgroups in the selection ofthe optimal total duration will allow for

greater precision in selecting andindividualizing regimen duration (57).

Drugs and Drug Classes

These guidelines are intended for settingsin which treatment is individualized based onDST results and clinical and epidemiologicalfactors. Individualized treatment regimensshould only include drugs to which thepatient’s isolate has documented or highlikelihood of susceptibility. Treatmentregimens should favor medications that areassociated with improved outcomes, asidentified in our PS-matched IPDMA,that limit toxicity and that incorporate patientpreferences. The following alphabeticallylisted drugs and drug classes were consideredfor inclusion in treatment regimens:amoxicillin/clavulanate, bedaquiline,carbapenem with clavulanic acid,clofazimine, cycloserine, delamanid,ethambutol, ethionamide, fluoroquinolones,injectable agents, linezolid, macrolides,p-aminosalicylic acid, and pyrazinamide.For each drug or drug class, the followingPICO was addressed: In patients with MDR-TB, are outcomes safely improved whenregimens include the following individualdrugs or drug classes compared withregimens that do not include them?

Amoxicillin/ClavulanateAmoxicillin–clavulanate, consisting of theb-lactam antibiotic amoxicillin and theb-lactamase inhibitor potassium clavulanate,is considered safe and effective fornumerous bacterial infections. AlthoughM. tuberculosis has an impenetrable cellwall and produces a b-lactamaseinhibitor (58), amoxicillin–clavulanatehas been used to treat TB. It is viewedas a “salvage” agent when few other

PICO Question 4—Amoxicillin/Clavulanate: In patients with MDR-TB, are outcomessafely improved when regimens include amoxicillin/clavulanate compared withregimens that do not include amoxicillin/clavulanate?

Recommendation 4: We recommend NOT including amoxicillin–clavulanate in atreatment regimen for patients with MDR-TB, with the exception of when the patientis receiving a carbapenem, wherein the inclusion of clavulanate is necessary (strongrecommendation, very low certainty in the evidence). Our recommendation againstthe use of amoxicillin–clavulanate (except to provide clavulanate when using acarbapenem) in MDR-TB treatment is strong despite the evidence being judged to beof very low certainty because we viewed the increased mortality and decreasedlikelihood of treatment success associated with the use of this drug as having a notablyunfavorable balance of benefits to potential harms.

Table 7. Distribution of Patients with Multidrug Resistance by Duration of Treatment from Sputum Culture Conversion to End ofTreatment (N=4,691)

Intervals from SputumCulture Conversion toEnd of Treatment (mo)

No. ofPatients

Mean Duration ofIntensive Phase

(mo)

Mean Duration toSputum CultureConversion (mo)

Mean Interval fromSputum Culture

Conversion to End ofTreatment (mo)

Total Duration ofTreatment (mo)

0.1–12.0 396 6.7 5.7 9.3 15.012.01–15.0 593 6.8 3.7 13.9 17.615.01–18.0 1,235 7.3 2.1 16.9 19.018.01–21.0 1,158 7.9 2.1 19.3 21.421.01–24.0 893 8.8 1.9 22.4 24.424.01–69 416 11.9 1.9 26.8 28.7

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e105

TB drug options remain availablebecause of either drug resistance orintolerance. The data for use ofamoxicillin–clavulanate in TB arelimited and show mixed results (59–62).In addition, because it is the only sourceof clavulanate, amoxicillin–clavulanatehas been used along with carbapenems.Synergy between meropenem andamoxicillin–clavulanate in XDR TBhas been noted (63, 64), and thecombination has been reported to beefficacious, safe, and tolerable whenadded to linezolid and other drugs(65–69). See PICO Question 6 for theevaluation of carbapenems withamoxicillin–clavulanate.

Summary of the evidence. Proceduresand methodology to assemble and thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B.Our PS-matched IPDMA showed thatpatients who received amoxicillin–clavulanate were more likely to have beentreated with second-line drugs and to beresistant to fluoroquinolones or anysecond-line injectable (3). They were morelikely to have received a later-generationfluoroquinolone (72%), capreomycin

(58%), linezolid (25%), and bedaquiline(8%). In adjusted analyses, patientswho received amoxicillin–clavulanatewere less likely to have treatment success(aOR, 0.6; 95% CI, 0.5–0.8) and morelikely to die (aOR, 1.7; 95% CI, 1.3–2.1)compared with patients who did not receiveamoxicillin–clavulanate.

Benefits. The addition of amoxicillin–clavulanate (without coadministration withcarbapenems) to a regimen for MDR-TBdoes not appear to provide benefit.

Harms. Patients who receivedamoxicillin–clavulanate were less likely tohave treatment success and more likely todie than patients who did not receiveamoxicillin–clavulanate. Data on adverseeffects were not collected systematicallyacross studies in our PS-matched IPDMAor in most trials. A published systematicreview identified diarrhea and candidiasisas key adverse effects associated withamoxicillin-clavulanic acid use (70).

Additional considerations. Clavulanicacid is only available as a coformulation withamoxicillin. Therefore, amoxicillin–clavulanatemust be given whenever carbapenemsare included in an MDR-TB regimen (seePICO Question 6 on carbapenems withclavulanate).

Conclusions. Patients who receivedamoxicillin–clavulanate were less likely toachieve treatment success and more likely todie than patients who did not receiveamoxicillin–clavulanate. These data suggestamoxicillin–clavulanate should not be used inMDR-TB treatment, except to provideclavulanate when using a carbapenem (seePICO Question 6 on carbapenems withclavulanate). Our recommendation againstthe use of amoxicillin–clavulanate (except toprovide clavulanate when using acarbapenem) in MDR-TB treatment is strongdespite the evidence being judged to be ofvery low certainty because we viewed theincreased mortality and decreased likelihoodof treatment success associated with the useof this drug as having a notably unfavorablebalance of benefits to potential harms.

Research needs. The development of aclavulanic acid formulation withoutamoxicillin, for use in combination withcarbapenems, would be helpful and wouldavoid unnecessary toxicities from amoxicillinas well as adhere to international efforts inpromoting antimicrobial stewardship (1).

BedaquilineBedaquiline, a diarylquinoline, approved byFDA in 2013, is the first drug with a novelmechanism of action againstM. tuberculosisto have been approved by FDA in .40years (71, 72). Bedaquiline is bactericidal tononreplicating and actively replicatingmycobacteria, through ATP synthaseinhibition, and has bactericidal andsterilizing activity in the murine model ofTB infection (73). No cross-resistance hasbeen found between bedaquiline and thefollowing: isoniazid, rifampin, ethambutol,pyrazinamide, streptomycin, amikacin, ormoxifloxacin. There have been a fewreports of cross-resistance with clofazimine

PICO Question 5—Bedaquiline: In patients with MDR-TB, are outcomes safelyimproved when regimens include bedaquiline compared with regimens that do notinclude bedaquiline?

Recommendation 5: We recommend including bedaquiline in a regimen for thetreatment of patients with MDR-TB (strong recommendation, very low certainty inthe evidence). Our recommendation for the use of bedaquiline is strong despite verylow certainty in the evidence because we viewed the significant reduction in mortality,improved treatment success, and relatively few adverse effects associated with MDR-TB treatment including bedaquiline (compared with no bedaquiline) as having aparticularly favorable balance of benefits over harms.

Table 8. Adjusted Estimates of Treatment Success by Duration of Treatment Interval between Sputum Culture Conversion and Endof Treatment, All Forms of Multidrug Resistance (N= 4,691)

Interval from Sputum CultureConversion to End of Treatment (mo)

No. of Patients Propensity Score–matched Analysis

TreatmentSuccess Total No. of Pairs aOR 95% CI

Risk Difference(95% CI)

0.1–12.0 360 396 394 0.5 0.4 to 0.7 20.04 (20.07 to 20.01)12.01–15.0 565 593 — 1.0 Reference —15.01–18.0 1,206 1,235 1,223 2.1 1.4 to 3.1 0.02 (0.01 to 0.04)18.01–21.0 1,122 1,158 1,154 1.6 1.1 to 2.3 0.02 (0.00 to 0.03)21.01–24.0 858 893 889 1.2 0.9 to 1.8 0.01 (20.01 to 0.02)24.01–69 386 416 413 0.7 0.4 to 1.0 20.02 (20.05 to 0.00)

Definition of abbreviations: aOR=adjusted odds ratio; CI = confidence interval.

AMERICAN THORACIC SOCIETY DOCUMENTS

e106 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

(74, 75). Bedaquiline is customarily used aspart of combination therapy (minimumfour-drug therapy) for adults aged >18years with a diagnosis of pulmonary MDR-TB when an effective treatment regimencannot otherwise be provided (e.g.,mycobacterial isolates show a complicateddrug resistance pattern, drug intolerance, ordrug–drug interactions) (76). Therecommended dose of bedaquiline for thetreatment of pulmonary MDR-TB in adultsis 400 mg administered orally once daily for2 weeks, followed by 200 mg administeredorally three times weekly, for an entiretreatment duration of 24 weeks (72, 76–78).Bedaquiline has recently has been identifiedas the key drug in assembling an all-oral,injectable-free drug regimen in SouthAfrica (79).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B. Our PS-matched IPDMA included 411 patients whoreceived bedaquiline-containing regimens, forwhom it was assumed that there was nobedaquiline resistance, and 10,932 patientswho did not receive bedaquiline (3). Patientswho received bedaquiline-containingregimens were more likely to have cavitarydisease (82% vs. 62%), to have been treatedwith second-line drugs (34% vs. 14%), to beinfected with an organism resistant tofluoroquinolones (57% vs. 21%) or to anysecond-line injectable (58% vs. 23%), andto have XDR-TB (29% vs. 13%). They werealso more likely to have received a later-generation fluoroquinolone (69% vs. 54%),linezolid (64% vs. 4%), or clofazimine (48%vs. 4%). Treatment success (cure andcompletion) was slightly greater (70% vs.60%; P=0.001) with bedaquiline, whereasfailure/relapse (6% vs. 9%), death (10% vs.15%), and loss to follow-up (14% vs. 16%)were less frequent in PS-adjusted analyses.

Benefits. Our PS-matched IPDMAshowed that patients treated withbedaquiline-containing regimens were morelikely to have treatment success (aOR, 2.0;95% CI, 1.4–2.9), less likely to experiencefailure/relapse, and less likely to die (aOR,0.4; 95% CI, 0.3–0.5; absolute riskreduction, 7.2%). Results were similar whenonly patients from high-income countriestreated with bedaquiline-containingregimens were included and whencomparing only patients with XDR-TB whowere and were not treated with bedaquiline.

A recent large program-based observationalmultinational study confirmed a hightreatment success rate (76.9%) withbedaquiline-containing regimens, with alow proportion (5.8%) of interruptionsowing to adverse events (80). In ourIPDMA, using PS-matched pairscomparing the effects of bedaquiline tothose of clofazimine, statistically significantdifferences for success versus failure/relapsefavoring bedaquiline use were observed(aOR, 2.1; 95% CI, 1.1–4.1) (using .170PS-matched pairs), but statisticallysignificant findings were not observed inmortality or when restricting analyses tohigh-income countries. PS-matched pairsanalyses comparing the effects of variouscombinations of drugs used withbedaquiline all noted improved outcomes.When comparisons of effects of bedaquilineand linezolid with those of no bedaquilineor linezolid were performed, combinationof bedaquiline and linezolid was associatedwith an aOR of 2.7 (95% CI, 1.5–4.9) forsuccess versus failure/relapse and with anaOR of 0.3 (95% CI, 0.2–0.4) for deathversus success/failure/relapse. Similarly,when PS-matched pair analyses comparingthe effects of bedaquiline and clofazimineto those of no bedaquiline or clofaziminewere performed, combination of bedaquilineand clofazimine was associated with an aORof 5.0 (95% CI, 2.4–10.6) for success versusfailure/relapse, and with an aOR of 0.3 (95%CI, 0.2–0.5) for death versussuccess/failure/relapse. Of note, patients towhom bedaquiline was administered tendedto have more cavitary disease or to have pre-XDR or XDR-TB. Prescription ofbedaquiline was associated with greatersuccess and less death than clofazimine, butthe greatest success and least mortality wasfound when bedaquiline was administeredtogether with linezolid or clofazimine.

Finally, a recent retrospective routine-care observational study in South Africashowed that a bedaquiline-containingregimen was associated with significantlylower mortality (128 [12.6%] deaths among1,016 patients receiving bedaquilinecompared with 4,612 deaths [24.8%]among 18,601 patients on the standardregimens). Bedaquiline was associated witha reduction in the risk of all-cause mortalityfor patients with MDR- or RR-TB(hazard ratio [HR], 0.35; 95% CI, 0.28–0.46)and XDR-TB (HR, 0.26; 95% CI,0.18–0.38) compared with standardregimens (50).

Harms. In a review of cases treatedwith bedaquiline, only 44 of 1,266 (3.5%)cases with information availablediscontinued bedaquiline because of adverseevents. Only 8 of 875 (0.9%) discontinuedbedaquiline because of QT intervalprolongation (two restarted the drug afterresolution of the acute episode withoutfurther problems) (81).

Additional considerations. Whenbedaquiline is included in the regimen, mostexperts obtain ECGs after the initial 2 weeksof therapy and then at monthly intervals tomonitor for QT interval prolongation.Serum electrolytes, including calcium,magnesium, and potassium, are alsomonitored.

The 2013 CDC guidelines on the use ofbedaquiline note that there is insufficientevidence to provide guidance on the use ofbedaquiline in children but that its use canbe considered on a case-by-case basis giventhe high mortality and limited treatmentoptions for MDR-TB (71). More recently,adolescents .10 years old and weight.34 kg have been safely treated off-labelwith the recommended adult dose ofbedaquiline (82). The Sentinel Project onPediatric Drug-Resistant Tuberculosis hasrecommended that children >12 years ofage and >31 kg body weight receivebedaquiline at the same dose as adults, andchildren >6 years with 16 to 30 kg bodyweight could receive half the adultbedaquiline dose for the same indications(7, 17, 83).

Conclusions. In our PS-matchedIPDMA, bedaquiline-containing regimenswere more likely to achieve treatmentsuccess and to have a lower rate of deaththan those that did not include bedaquiline.Bedaquiline should be included in a regimento achieve a total of five effective drugs forthe treatment of patients with MDR-TB.Our recommendation for the use ofbedaquiline is strong despite very lowcertainty in the evidence because we viewedthe significant reduction in mortality,improved treatment success, and relativelyfew adverse effects associated with MDR-TBtreatment including bedaquiline (comparedwith no bedaquiline) as having a particularlyfavorable balance of benefits over harms.

Research needs. Further research isneeded to elucidate the potential synergy ofbedaquiline with other agents. Studiesunderway are evaluating the use ofbedaquiline together with linezolid,clofazimine, or nitroimidazoles

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e107

(i.e., pretomanid and delamanid). Research isalso needed on the safety, tolerability, andefficacy of bedaquiline-based shorter-courseregimens as well as on the use of bedaquilinefor durations .24 weeks, an approachcurrently considered when effective treatmentcannot otherwise be provided (71, 84).Finally, research is urgently needed on therisk factors and any interventions (e.g., theselection of companion drugs) that influenceacquisition of bedaquiline resistance.

Carbapenems with Clavulanic AcidThe combination of carbapenems andclavulanate, a b-lactamase inhibitor, hasbeen shown to have in vitro bactericidalactivity (65, 85, 86). As clavulanate isnot available by itself, the combinationdrug amoxicillin–clavulanate must begiven with the carbapenems. Carbapenemshave been used primarily for MDR andXDR TB, and a recent systematic reviewfound that carbapenems are safe andlikely to be effective (65). Carbapenemdrugs have been recently included inWHO guidelines for the treatment ofDR-TB (23).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B.Our PS-matched IPDMA compared death,treatment success, and culture conversionamong 169 individuals who receivedcarbapenems to 9,535 individuals fromcenters that did not use any of the drugspreviously classified by WHO as “Group 5”drugs (3). In addition, we consideredadverse events reported in a recent review(65), IPDMA, and systematic review thatsummarize the available evidence oncarbapenems from five primary studies (all

observational studies) (23, 65–67, 87–92).The reviewed evidence showed that therewas concomitant use of linezolid andbedaquiline in large proportions of thepatient population receiving carbapenems,suggestive of confounding, which wasconsidered in the review of the efficacyresults.

Benefits. In our PS-matched IPDMA,inclusion of carbapenems in the treatmentregimen had no effect on the risk ofdeath (aOR, 1.0; 95% CI, 0.5–1.7) orculture conversion (aOR, 2.3; 95% CI,0.8–6.9). However, treatment success wasmore likely among patients treated withregimens including a carbapenemcompared with those treated with regimensnot including a carbapenem (aOR, 4.0; 95%CI, 1.7–9.1).

Harms. In our meta-analysis, the rateof treatment discontinuation due to anadverse event was lower among patientstreated with regimens including acarbapenem than those treated withregimens not including a carbapenem,although this difference was not statisticallysignificant (relative risk, 0.82; 95% CI,0.23–2.94). In the published literature,treatment discontinuation due to adverseevents has been reported in 0 to 3% ofpatients, with minor adverse eventsoccurring in 5% to 6% of patients (65).

Additional considerations. Of note,clavulanic acid is only available as acoformulation with amoxicillin. Therefore,amoxicillin–clavulanate must be given toachieve a dose of 125 mg of the clavulanatewith each dose of carbapenem daily,whenever carbapenems are included in anMDR-TB regimen. DST is currently notavailable for carbapenems. When used,imipenem–cilastatin/clavulanate ormeropenem/clavulanate have beenadministered intravenously in a hospitalsetting, with multiple injections requireddaily (67). There is no clear evidence ofwhether imipenem–cilastatin ormeropenem is more efficacious (66).Ertapenem belongs to the same family(65, 93); because of its longer half-life andthe possibility to administer it once aday intramuscularly or intravenously,ertapenem may be useful when a patienttreated with meropenem orimipenem–cilastatin intravenously duringhospitalization is discharged and needs tocontinue carbapenem-based treatmentoutpatient (92). Long-term intravenousadministration requires long-term venous

access through an indwelling catheter,which carries risks of infection,thrombosis, and thromboembolism. Dataon the use of a carbapenem combined withamoxicillin–clavulanate in children arelacking—only two adolescents areincluded in one review (65). However,efficacy has been shown in adults, andboth carbapenems and amoxicillin–clavulanate have been used safely inchildren for other purposes (althoughthere are no long-term treatmentsafety data in young children); therefore,this combination could be used inchildren with MDR-TB where there isno other option to build an effectiveregimen.

Conclusions. In our PS-matchedIPDMA, inclusion of carbapenems withclavulanic acid was associated with anincrease in treatment success whencompared with the control group notreceiving carbapenems with clavulanic acid.Carbapenems with clavulanic acid can beincluded in a regimen to achieve a total offive effective drugs for the treatment ofpatients with MDR-TB.

Research needs. Randomized,controlled clinical trials that confirm therole of carbapenems used for differentdurations of time within differentregimens are needed. A comparativeevaluation of the safety, tolerability,and efficacy of the different agents isalso necessary. Because of the high costof these drugs, economic analyses willalso be useful (65). The developmentof oral formulations of carbapenemsis underway, which, if proven safe andeffective, would enhance feasibility ofusing these agents significantly.Additional advances in rapid DSTfor carbapenems would enhancepotential scale-up of use for this classof drugs.

PICO Question 7—Clofazimine: Inpatients with MDR-TB, are outcomessafely improved when regimensinclude clofazimine compared withregimens that do not includeclofazimine?

Recommendation 7: We suggestincluding clofazimine in a regimen fortreatment of patients with MDR-TB(conditional recommendation, verylow certainty in the evidence).

PICO Question 6—Carbapenems withclavulanic acid: In patients with MDR-TB, are outcomes safely improvedwhen regimens include carbapenemswith clavulanic acid compared withregimens that do not include them?

Recommendation 6: We suggestincluding a carbapenem (always to beused with amoxicillin–clavulanic acid)in a regimen for treatment of patientswith MDR-TB (conditionalrecommendation, very low certainty inthe evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e108 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

ClofazimineClofazimine, a fat-soluble riminophenazinedye, has shown in vitro and in vivo activityas a sterilizing drug to treat MDR-TB(94–96).

Clofazimine has been used as a leprosydrug and was first developed in the 1950sbecause of in vitro and in vivo activity againstM. tuberculosis (94, 95, 97). Although theexact mechanism of action is not known, it isa prodrug that appears to have bothantimycobacterial and antiinflammatoryproperties (98). The published clinicalevidence on the safety and efficacy ofclofazimine for treatment of TB is modest(99). However, interest in the drug hasincreased since WHO endorsed the newshorter-course regimen (23, 100), whichincludes clofazimine (101–108).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B. Our PS-matched IPDMA compared death, treatmentsuccess, and culture conversion among 639individuals who received clofazimine to 7,398individuals from centers that did not use anyof the drugs previously classified by WHOas “Group 5” drugs (3). In addition, weconsidered adverse events reported in apublished systematic review of nineobservational studies (six MDR-TB and threeXDR-TB) including 599 patients treated withclofazimine (427 from a single cohort studyfrom Bangladesh) (95) and two studiespublished after the systematic review. Theyinclude a trial at six specialty hospitals inChina that randomized 105 patients to 21months of treatment with an optimizedbackground regimen with or withoutclofazimine (100 mg/d) (94) and aretrospective cohort study from Brazilthat compared outcomes in patients withMDR-TB treated with clofazimine(100 mg/d)-containing regimens versuspyrazinamide-containing regimens (109).

Benefits. Our PS-matched IPDMAshowed that treatment success was more likelywith regimens containing clofaziminecomparedwith regimens that did not (aOR, 1.5;95% CI, 1.1–2.1). Addition of clofazimine wasassociated with an aOR of 0.8 (95% CI,0.6–1.0) for the outcome of death and with anaOR of 1.1 (95% CI, 0.6–1.8) for cultureconversion within 6 months. As noted inPICO Question 5 on bedaquiline, when PS-matched pair analyses were performed

comparing the effects of bedaquiline andclofazimine to those of no bedaquiline orclofazimine, the combination of bedaquilineand clofazimine was associated with an aOR of5.0 (95% CI, 2.4–10.6) for success versusfailure/relapse and with an aOR of 0.3 (95% CI,0.2–0.5) for death versus success/failure/relapse.

Harms. In our PS-matched IPDMA, 2of 81 (2.5%) patients treated with regimensincluding clofazimine had treatmentdiscontinued because of adverse events.Brownish skin pigmentation has beenobserved in 75% to 100%, ichthyosis in 8%to 20%, gastrointestinal intolerance in 40%to 50%, and neurological disturbances inup to 13% of patients (94, 95, 109).Clofazimine was judged to have small tomoderate desirable effects (on treatmentsuccess, mortality, and culture conversions)and small undesirable effects (low risk ofserious adverse events requiring treatmentdiscontinuation). However, there isimportant uncertainty about how patientsvalue skin discoloration. Some patientsperceive skin discoloration as significant,which may become a key factor in theacceptability of this drug in some settings.Recent publications have noted thepotential QT interval–prolonging effects ofclofazimine when used in combination withbedaquiline and delamanid (110–112).

Additional considerations. Access is alimitation to expanded use of clofazimine fortreatment of MDR-TB, especially in the UnitedStates and Europe, where clofazimine lacks aTB indication. In the United States, clofazimineis currently only available under aninvestigational new drug protocol administeredby FDA, which can be a burdensomeprocedure. Significant improvements in themechanisms for accessing clofazimine arenecessary for more expanded use of this drug.Quality-assured clofazimine is availablethrough the Global Drug Facility, althoughcurrently in limited quantities. Although ourPS-matched IPDMAhad limited pediatric data,the efficacy in children is believed by experts tobe the same as in adults. A recent IPDMA inchildren with MDR-TB did not show a benefitusing clofazimine, which might be related toselection of cases and very small numbers (23 of641) receiving clofazimine (113). Moreover,children are included in the WHO shorter-course regimen for MDR-TB treatment, whichincludes clofazimine, albeit on the basis oflimited data. Dosing clofazimine in children ischallenging because of the lack of child-friendly formulations and lack ofpharmacokinetic data. The currently

recommended dose for clofazimine in childrenvaries from 2 to 5 mg/kg daily. Skindiscoloration is common; some improveduring treatment and others resolve rapidlyafter discontinuation of treatment. Ichthyosisis less common and improves with intensiveefforts at applying lubricants during treatment.As noted, QT interval prolongation may occurand is especially important to monitor by ECGif given together with other QTinterval–prolonging medication (68, 83,110–112, 114, 115). Clofazimine may havecross-resistance with bedaquiline, which mayneed to be considered when building aregimen for MDR-TB (75).

Conclusions. In our PS-matchedIPDMA, inclusion of clofazimine wasassociated with increase in treatmentsuccess when compared with the controlgroup not receiving clofazimine.Clofazimine can be included in a regimen toachieve a total of five effective drugs for thetreatment of patients with MDR-TB.

Research needs. The value of usingloading doses and the optimal dosing forclofazimine requires additional research.Pharmacokinetic data in adults and childrenare needed.

CycloserineCycloserine is an oral bacteriostatic drug thathas been part of the backbone regimen forMDR-TB treatment in the past. Cycloserine isa broad-spectrum antibiotic that inhibits cellwall synthesis (116, 117). Terizidone is astructural analog that is a combination of twocycloserine molecules; it appears to be usedinterchangeably by experts, although it iscurrently not available in the United States(118). Although some advantages ofcycloserine include the absence of cross-resistance to other drugs and reasonablegastrointestinal tolerability, a significantdrawback relates to psychological side effects,which occasionally necessitatediscontinuation of the drug (16).

Summary of the evidence. Previousstudies showed an increase of the likelihood oftreatment success when cycloserine wasincluded in theMDR-TB regimen (marginallystatistically significant) (96). Procedures andmethodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B.

Benefits. Our PS-matched IPDMAshowed that inclusion of cycloserine wasassociated with an increase in treatment success(aOR, 1.5; 95% CI, 1.4–1.7) and a decrease in

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e109

mortality (aOR, 0.6; 95% CI, 0.5–0.6) whencompared with the control group. A publishedIPDMA of children with successful treatmentof MDR-TB showed that cycloserine as a singledrug had an aOR of 1.7 (95% CI, 0.9–3.0; 641children with MDR-TB, of whom 356 [56%]received cycloserine/terizidone) (113). Limitedpharmacokinetic data are available oncycloserine (or terizidone) in children, but arecent study of 25 children (only 5 to ,12 yrof age) receiving a median of 14.3 mg/kg(range, 10.0–18.0 mg/kg) showed that themaximum concentration of the drug achievedin the plasma after dose administration wassimilar to the adult target maximumconcentration of 20 to 35 mg/L (119). Therecommended dose for children of 10 to 20mg/kg/d seems adequate. An advantage ofcycloserine in children, as in adults, withmiliary and/or central nervous system (CNS)TB is that it penetrates the CNS well.Cycloserine may interfere with thepharmacokinetics and absorption of isoniazidand ethionamide/prothionamide; therefore, itmay be advisable to give it separately fromthese drugs if used together in the sameregimen (117).

Harms. Cycloserine has CNS adverseeffects, which are reported to occur in about20% to 30% of adults (120). A meta-analysisof published articles identified adverseevents in 201 of 2,164 patients across allstudies (118). An average weightedproportion of patients receiving cycloserinewho discontinued treatment owing to adverseeffect pooled across the studies was 9.1%(95% CI, 6.4–11.7%). An average weightedproportion of patients with psychiatricadverse effects was 5.7% (95% CI, 3.7–7.6%)(118). Limited data exist on cycloserineadverse effects in children. In older pediatricstudies, no adverse effects were reported withthe use of cycloserine (121–123). In asystematic review of outcomes of MDR-TB inchildren, 6 of 182 (3.3%) children had adverse

effects attributed to cycloserine, whichincluded depression, anxiety, hallucinations,transitory psychosis, and blurred vision (124).

Additional considerations. Lower-than-recommended serum cycloserineconcentrations and delayed absorption havebeen reported (125). Some experts obtainpeak concentrations within the first 1 to 2weeks of therapy and continue to monitorserially. Because of specific technicalchallenges related to cycloserine DST, pooraccuracy of testing in liquid media, and poorintrinsic reproducibility of results, fewlaboratories perform DST for cycloserine (16).

Conclusions. In our PS-matchedIPDMA, inclusion of cycloserine wasassociated with a decrease in mortality andincrease in treatment success whencompared with the control group notreceiving cycloserine. Cycloserine can beincluded in a regimen to achieve a total offive effective drugs for the treatment ofpatients with MDR-TB.

Research needs. Studies are needed todetermine the effect of cycloserine on theabsorption of isoniazid and ethionamide.

DelamanidDelamanid is a nitro-dihydro-imidazooxazolederivative, which was approved for thetreatment of MDR-TB by the EuropeanMedicines Agency in 2013 but has not yetreceived FDA approval. In 2014, WHO issuedinterim policy guidance on the use ofdelamanid for the treatment of MDR-TB onthe basis of phase 2b clinical trial data. Theinterim policy guidance stated that“delamanid may be added to an MDR-TBregimen in adult patients with pulmonary TBconditional upon: i) careful selection ofpatients likely to benefit; ii) patient informedconsent; iii) adherence to WHOrecommendations in designing a longerMDR-TB regimen; iv) close monitoring of

clinical treatment response; and v) active TBdrug-safety monitoring and management(aDSM)” (126). WHO issued an updatedposition statement on the use of delamanidfor MDR-TB in 2018 on the basis of finalresults from the phase 3 randomizedcontrolled trial, Trial 213 (127).

Summary of the evidence. Delamaniddata were not available as part of thePS-matched IPDMA completed for thisguideline development process; however,the 2019 WHO Consolidated Guidelines onDrug-Resistant Tuberculosis Treatment,wherein individual-level data on delamanidwere attained for analyses, provides practiceguidance on the use of delamanid fortreatment of MDR-TB (7, 128). Severalrecent reports, clinical trials, and cohortstudies report on delamanid-containingregimens achieving success rates of 77% to84%, in absence of major adverse events(128–131). In the United States, through acompassionate use program, delamanid hasrecently been successfully accessed andused in the care of patients with MDR-TB(132). Regarding use of delamanid inchildren, only a small number of children(.6 yr of age) have had documented accessto the drug, also by compassionate use (133,134). On the basis of review of the dataavailable in the IPDMA conducted byWHO, delamanid has been included in“Group C” in their guidelines,corresponding to drugs that can be added“to complete the regimen and whenmedicines from Groups A and B cannot beused.” WHO has recommended that“delamanid may be included in thetreatment of MDR/RR-TB patients aged 3years or more on longer regimens” (7, 135).The guideline writing committee concurswith the updated 2019 WHO guidance(7). As noted for other drugs discussed inthis practice guideline, QT interval

PICOQuestion 9—Delamanid: In patients withMDR-TB, are outcomes safely improvedwhen regimens include delamanid compared with regimens that do not includedelamanid?

Recommendation 9: The guideline panel was unable to make a clinicalrecommendation for or against delamanid because of the absence of data in the PS-matched IPDMA conducted for this practice guideline. We make a researchrecommendation for the conduct of randomized clinical trials and cohort studiesevaluating the efficacy, safety, and tolerability of delamanid in combination with otheroral agents. Until additional data are available, the guideline panel concurs with theconditional recommendation of the 2019 WHO Consolidated Guidelines on Drug-Resistant Tuberculosis Treatment that delamanid may be included in the treatment ofpatients with MDR/RR-TB aged >3 years on longer regimens (7).

PICO Question 8—Cycloserine: Inpatients with MDR-TB, are outcomessafely improved when regimensinclude cycloserine compared withregimens that do not includecycloserine?

Recommendation 8: We suggestincluding cycloserine in a regimen fortreatment of patients with MDR-TB(conditional recommendation, verylow certainty in the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e110 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

prolongation may occur with delamaniduse, and routine ECG monitoring isrecommended. A recent evaluation of thecardiac safety of delamanid and bedaquilinegiven together as part of multidrug therapyfor MDR-TB concluded that the combinedeffect on the corrected QT interval (QTc)using the Fridericia formula (QTcF) isclinically modest and no more thanadditive (ClinicalTrials.gov Identifier:NCT02583048), with mean change in QTcFfrom baseline at 11.9 milliseconds (95.1%CI, 7.4–16.5 ms) in the bedaquiline arm, 8.6milliseconds (95.1% CI, 4.0–13.2 ms) in thedelamanid arm, and 20.7 milliseconds(95.1% CI, 16.1–25.4 ms) in the combinedbedaquiline and delamanid arm (136).

Conclusions. The guideline panel wasunable to make a recommendation for oragainst delamanid because of the absence ofdata in the IPDMA conducted for thispractice guideline. In 2018, WHO updatedthe IPDMA with additional data andrecommended delamanid be included in thethird tier of drugs, Group C, and for the drugto be used in the treatment of patients withMDR/RR-TB aged >3 years on longerregimens. The guideline writing committeeagrees with the updated 2019 WHOConsolidated Guidelines on Drug-ResistantTuberculosis Treatment that delamanidmay be included in the treatment ofpatients with MDR/RR-TB aged >3 yearson longer regimens (7). We make aresearch recommendation for the conductof randomized clinical trials and cohortstudies evaluating the efficacy, safety, andtolerability of all-oral, shortened regimensinclusive of delamanid in combination withother oral agents.

Research needs. The companion drugswith which to combine delamanid for

optimal efficacy, safety, and tolerabilityremain uncertain. The conduct ofrandomized clinical trials and cohort studiesevaluating all-oral, shortened regimens,inclusive of delamanid and in combinationwith other oral agents are urgently needed.The endTB trial is a phase 3 multicountryrandomized clinical trial evaluating theefficacy and safety of five new, all-oral,shortened regimens including combinationsof bedaquiline and delamanid(ClinicalTrials.gov Identifier: NCT02754765).Recent cohort data are also emerging on theuse of delamanid and bedaquiline incombination, which to date has beengenerally reserved for severe cases withextensive resistance and when other optionsare not feasible (137–140). The Nix-TB (APhase 3 Study Assessing the Safety andEfficacy of Bedaquiline Plus PA-824 PlusLinezolid in Subjects with Drug-ResistantPulmonary Tuberculosis) trial(ClinicalTrials.gov Identifier: NCT02333799)evaluated an all-oral 6-month regimencomprising bedaquiline, pretomanid(member of the nitroimidazooxazines class ofcompounds), and linezolid for the treatmentof either XDR-TB or treatment-intolerant ornonresponsive MDR-TB. FDA grantedpriority review of the new drug applicationfor pretomanid in March 2019, theAntimicrobial Drugs Advisory Committeediscussed pretomanid in June 2019, and inAugust 2019 FDA approved pretomanid incombination with bedaquiline and linezolidfor the treatment of a specific limitedpopulation of adults with pulmonary XDR-TB or treatment-intolerant or nonresponsiveMDR-TB (2, 141). Given FDA approval,additional trials that include diverse patientpopulations will be essential to understandoptimal use of the regimen, in addition tohead-to-head comparisons of pretomanidwith delamanid to determine if these drugscan be used interchangeably.

EthambutolEthambutol is an ethylenediamine thatinhibits arabinosyl transferases, whichcontribute to M. tuberculosis cell wallsynthesis (142). The drug is included instandard regimens for treatment of drug-susceptible TB and commonly used inregimens for MDR-TB (11, 16, 23). Thepublished evidence available on its safetyand efficacy is modest, and its use as part ofthe first-line drug-susceptible TB regimenrests largely on its ability to prevent the

emergence of resistance to the other drugsin the regimen rather than on its ownsterilizing activity (143). Ethambutol hasbeen demonstrated to have modeststerilizing activity as part of combinationregimens when used against ethambutol-susceptible isolates for treatment of drug-susceptible TB. The drug is not effectivewhen used for ethambutol-resistant isolates,and such use is not recommended (54, 96).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B. OurPS-matched IPDMA compared outcomesamong 3,002 individuals with isolatessusceptible to ethambutol who received thedrug to 667 individuals with isolates susceptibleto ethambutol who did not receive the drug(3). Receipt of ethambutol was associated withan aOR of 0.9 (95% CI, 0.7–1.1) for theoutcome of treatment success; receipt ofethambutol was associated with an aOR of 1.0(95% CI, 0.9–1.2) for the outcome of death.However, the groups were not balanced withregard to other effective drugs; in particular,the patients not receiving ethambutol weremore likely to receive linezolid. Thus, theability to detect an effect of ethambutol mighthave been obscured. In a previous IPDMA,before linezolid and bedaquiline use werecommon, administration of ethambutol amongpatients with susceptible isolates was associatedwith an aOR of 1.7 (95% CI, 1.2–2.4) forcure/complete versus failure/relapse and anaOR of 1.6 (95% CI, 1.1–2.4) for cure/completeversus failure/relapse/death, compared withpatients receiving ethambutol whose isolateswere resistant in vitro (144).

Benefits. Our PS-matched IPDMAshowed that treatment success was notsignificantly more likely with regimenscontaining ethambutol. However, a numberof earlier studies indicate that ethambutol isassociated with increased success when usedto treat patients whose isolates aresusceptible to the drug (143). Moreover,these earlier studies determined that theeffectiveness of ethambutol is directlyrelated to dose, and that a dose of 25 mg/kgwas more effective than a dose of 15 mg/kg(143). Dosages were not available foranalysis in our PS-matched IPDMA.Finally, our PS-matched IPDMA did notaddress a key attribute of ethambutol, theprevention of the emergence of resistance,which is a substantial concern amongpatients with MDR-TB (145).

PICO Question 10—Ethambutol: Inpatients with MDR-TB, are outcomessafely improved when regimensinclude ethambutol compared withregimens that do not includeethambutol?

Recommendation 10: We suggestincluding ethambutol in a regimen fortreatment of patients with MDR-TBonly when more effective drugs cannotbe assembled to achieve a total of fiveeffective drugs in the regimen(conditional recommendation, verylow certainty in the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e111

Harms. In a recent review of thetolerability of TB drugs, ethambutol wasassociated with serious adverse events in 6 of1,325 (0.5%) patients (23). This is consistentwith previous reports of ethambutol toxicity,in which optic neuropathy (including opticneuritis and retrobulbar neuritis) wasattributed to ethambutol, manifesting assymptoms of decreased visual acuity,scotomata, color blindness, or visual defects.These toxic effects are dose dependent and,although serious, are generally reversible ifrecognized promptly and the drug isdiscontinued or the dose reduced.

Additional considerations. Whenethambutol is used, some experts recommendusing the higher dose of 25 mg/kg, which isassociated with increased efficacy but alsoslightly greater ocular toxicity. Other expertsrecommend using 15 to 20mg/kg, counting thedrug predominantly as providing protectionagainst the acquisition of additional resistance.All patients receiving ethambutol as part of anMDR-TB treatment regimen should bemonitored monthly for signs of ocular toxicity,particularly visual impairment, and if this isdetected, ethambutol should be discontinued. Ifoptic neuritis occurs in patients taking bothethambutol and linezolid, both drugs must bestopped. Many patients may be rechallengedsuccessfully with linezolid once visionnormalizes, but rechallenge with ethambutol isnot recommended. The efficacy of ethambutolis not expected to be different in childrencompared with adults. A dose of 25 mg/kg isrecommended by experts for children withMDR-TB. Two reviews on safety of ethambutolrelating to vision have shown low risk of oculartoxicity (143, 146). However, other drugs, suchas linezolid, may also cause optic neuritis, andthis should be taken into consideration whentreating MDR-TB in children, especially asscreening for visual loss is difficult inyoung children.

Conclusions. We suggest includingethambutol in a regimen for treatment ofpatients with MDR-TB only when moreeffective drugs cannot be used to achieve atotal of five effective drugs in the regimen(conditional recommendation, very lowcertainty in the evidence). The committee wasdivided on the value of recommendingethambutol, given the drug was not associatedwith any benefit in terms of treatmentsuccess or mortality. Concerns about thecomparability of IPDMA patient groupstreated with and without ethambutol werenoted. Unlike ethionamide/prothionamide,where there are substantial undesirable effects,

ethambutol had small undesirable effects(low risk of serious adverse events requiringtreatment discontinuation). Finally, ourPS-matched IPDMA did not address theprevention of emergence of resistance—arecognized attribute of ethambutol in thetreatment of drug-susceptible TB—althoughthe applicability of this attribute to MDR-TBregimens is unknown. Overall, the committeesuggested that ethambutol can be used onlywhen more effective drugs cannot beassembled to achieve the necessary fiveeffective drugs in the regimen.

Research needs. More evidence isneeded on the beneficial role that ethambutolmay play in preventing emergence of drugresistance in MDR-TB, including to neweroral drugs. More research is also needed toimprove on the reliability of DST forethambutol in different settings.

Ethionamide and ProthionamideEthionamide and prothionamide arederivatives of isonicotinic acid, somewhatsimilar in structure to isoniazid. Ethionamideis a prodrug and requires activation, afterwhich it inhibits mycobacterial fatty acidsynthesis that is necessary for cell wallsynthesis and repair. Clinical studies indicateefficacy of these drugs, and they have beenincluded in regimens for treatment of MDR-TB and for treatment of TB meningitis inadults and children (16, 23, 147, 148).

Summary of evidence. Procedures andmethodology to assemble and rank the certaintyin the evidence are reported in APPENDIX A, withevidence profiles for PICO questions reportedin APPENDIX B. Our PS-matched IPDMAshowed that ethionamide/prothionamide werenot associated with any benefit, even ifsusceptible by phenotypic DST (3). Previousstudies have shown, however, an increase inthe likelihood of treatment success ifethionamide was included in the MDR-TBregimen (23, 149). These different results canbe explained by the inclusion of newer andmore effective drugs, improving outcomes inthe comparator groups, potentially biasing theobserved effect against ethionamide; fewerpatients in the ethionamide/prothionamidegroup, compared with control subjects,received newer-generation fluoroquinolone(51% vs. 76%), amikacin (17% vs. 35%), andlinezolid (5% vs. 18%), and more patient in theethionamide/prothionamide groups receivedkanamycin (51% vs. 12%) and capreomycin(27% vs. 16%).

Benefits. No benefits were identified forinclusion of ethionamide/prothionamide in

regimens, even if susceptible by phenotypicDST. A pediatric IPDMA of MDR-TB casesalso did not show benefit using ethionamide/prothionamide; however, the majority ofchildren (590 of 641) in this study did receiveethionamide or prothionamide, potentiallylimiting the ability to discern a benefit (113).

Harms. The potential of ethionamideto cause adverse events can also limit itstolerability (120). A review of studiescomparing a regimen containingethionamide or prothionamide (all studiesbefore 1970) showed that adverse effectsleading to discontinuation of treatment wereequally frequent with ethionamide (11.3%;range 6–42%) and prothionamide (11.9%;range 6–40%). Adverse effects, whenreported, included abnormal liver functiontests, gastrointestinal intolerance, endocrinedysfunction, and hypothyroidism, the latteroccasionally requiring treatment withthyroxine (147). Hypothyroidism is acommon adverse effect of ethionamide(experienced by approximately 20%) (150,151) and is particularly noted if used withp-aminosalicylic acid and in HIV-infectedchildren (152). Gastrointestinal disturbance iscommon but usually resolve within the first2 weeks of treatment in children.

Additional considerations. Whenethionamide/prothionamide is included in aregimen for which five other effective drugscannot be assembled, experts recommenddose escalation (drug ramping) at time oftreatment initiation, as well as monitoring ofthyroid-stimulating hormone for evidenceof hypothyroidism requiring replacement(16). Of note, in the presence of inhAmutation, many isolates will show

PICO Question 11—Ethionamide/prothionamide: In patients with MDR-TB, are outcomes safely improvedwhen regimens include ethionamide/prothionamide compared withregimens that do not includeethionamide/prothionamide?

Recommendation 11: We suggestNOT including ethionamide/prothionamide in a treatment regimenfor patients with MDR-TB if newerand more effective drugs are availableto construct a regimen with at leastfive effective drugs (conditionalrecommendation, very low certaintyin the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e112 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

cross-resistance between isoniazid andethionamide/prothionamide (153).

Conclusions. When the individualizedtreatment regimen for patients withMDR-TB contains newer-generation, more-effective drugs, the addition ofethionamide/prothionamide does notappear to provide benefit.

Research needs. Research efforts areunderway to identify potential boosters ofpotency for ethionamide, which, ifsuccessful when coadministered, mayresult in improved therapeutic index andoverall better risk-benefit ratio for use (154).

Fluoroquinolones: Levofloxacin,Moxifloxacin, Ciprofloxacin, andOfloxacinThe fluoroquinolones are a family ofchemically related drugs characterized by acommon core dual-ring structure (155).Ofloxacin, then levofloxacin, then moxifloxacinsequentially improved on earlier generation’sspectrum of activity, including mycobacteria,and their antimycobacterial action increased asevidenced by lower minimum inhibitoryconcentrations (MICs) and increasing successin clinical use (156, 157). Physicians beganusing these drugs to treat MDR-TB on thebasis of in vitro data, with subsequent caseseries and observational studies showingefficacy (158), although none of thefluoroquinolones are currently indicated byregulatory authorities for the treatment ofTB. In general, these drugs are well absorbedorally, have favorable pharmacologicalprofiles for once-daily dosing, are generallywell tolerated, and now are available ingeneric formulations (155).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B. In our PS-matched IPDMA, ofloxacin was used mostcommonly (n=4,020), followed by levofloxacin(n=3,872), moxifloxacin (n=2,132), andciprofloxacin (n=431); 734 patients did notreceive a fluoroquinolone, the comparisongroup for subsequent analyses, and 828 patientsreceived two or more quinolones and wereexcluded (3). The groups were similar in age,sex, proportion AFB smear positive, proportionwith cavitary disease on chest radiograph, andprior treatment with first-line drugs. However,the no-quinolone group had substantially moreHIV coinfection (44% vs. 13–28%; proportionwith HIV coinfection across different

quinolones), more past treatment with second-line drugs (52% vs. 8–25%), more quinoloneresistance (86% vs. 6–33%), more second lineinjectables resistance (74% vs.12–34%), andmore XDR TB (73% vs. 4–19%). In terms oftreatment, the no-fluoroquinolone groupreceived less amikacin (7% vs. 18–43%) andkanamycin (13% vs, 26–65%) and morecapreomycin (66% vs. 6–34%). Therefore, themedian number of effective drugs (intensivephase) was lower in the no-fluoroquinolonegroup (2.1) than in the other groups (3.3–4.0).Thus, treatment success was lower in the no-fluoroquinolone group (35% vs. 55–68%) andfailure/relapse was higher than but not greatlydifferent from the others (13% vs. 2–10%);mortality, however, was much higher (40% vs.11–15%).

Benefits. In our PS-matched IPDMA,among patients with susceptible isolates,levofloxacin-containing regimens comparedwith no quinolone were associated withsignificantly more treatment successes(aOR, 4.2; 95% CI, 3.3–5.4) and significantlyfewer deaths (aOR, 0.6; 95% CI, 0.5–0.7).Moxifloxacin, compared with noquinolone, was also associated withsignificantly more treatment successes(aOR, 3.8; 95% CI, 2.8–5.2) andsignificantly fewer deaths (aOR, 0.5; 95%CI, 0.4–0.6). In the subgroup withresistance to an injectable drug(s),levofloxacin or moxifloxacin wereassociated with a significant improvementin treatment success (aOR, 1.8; 95% CI,1.2–2.8) and reduction in death (aOR, 0.6;95% CI, 0.4–0.8), although thecorresponding adjusted risk differenceswere not statistically significant. In pairwisecomparisons, both levofloxacin andmoxifloxacin were associated withsignificantly better treatment outcomesthan ofloxacin. The aORs of death werelower for the two later-generationquinolones when compared with ofloxacin(levofloxacin: aOR, 0.8; 95% CI, 0.6–0.9;moxifloxacin: aOR, 0.8; 95% CI, 0.6–1.0)(data not shown). Ofloxacin andciprofloxacin are considered inferiorquinolones against M. tuberculosis (144,149, 159, 160). Levofloxacin andmoxifloxacin did not differ significantlyfrom each other.

In a recent IPDMA describingtreatment outcomes in children treated forMDR-TB (113), new and repurposed TBdrugs, including late-generationfluoroquinolones, were not used enough toadequately evaluate efficacy, but most experts

and emerging evidence suggest that theefficacy of fluoroquinolones noted in adultsshould be similar in children (83, 161).

Harms. In our PS-matched IPDMA,grade 3 adverse events were recordedsystematically in a subset of 1,962 patientsfrom a cohort study of MDR-TB at27 sites in nine countries. Permanentdiscontinuation of fluoroquinolones due toadverse events was uncommon. Among 150patients treated with levofloxacin, the drugwas stopped permanently because of adverseevents in 6 (4.0%). Among 398 patientstreated with moxifloxacin, the drug wasstopped permanently in 14 (3.5%) patients.Among 1,167 patients treated withofloxacin, the drug was stoppedpermanently in 56 (4.8%) patients. Ananalysis of 56 clinical trials comparingquinolones against placebo or against otherantimicrobial agents found generally similaradverse event profiles (3). Seven studiesreported more frequent adverse events andsix studies reported fewer adverse events influoroquinolone-treated patients. The mostfrequent adverse effects reported are thoseof the gastrointestinal tract in 3% to 17%and CNS in 0.9% to 11% of patients (155,162). Allergic and other hypersensitivityreactions and other skin reactions occur in0.5% to 2.8% of patients. Other adverseeffects that occur in .1% of patients arecardiac (QT interval prolongation) andendocrine (hypoglycemia). Recently, FDAstrengthened warnings in the prescribinginformation for the entire class offluoroquinolones on risks of severehypoglycemia, certain mental health sideeffects, and tendonitis, as well as risksof ruptures or tears in the aorta (163).Safety concerns persist for long-termpediatric use of fluoroquinolones,especially regarding arthropathy.However, several long-term prospectiveand retrospective studies in children haveconfirmed that severe adverse effects withthe fluoroquinolones are rare, includingmusculoskeletal, neurological, and QTinterval prolongation adverse effects(164–166).

Additional considerations. As later-generation fluoroquinolones have becomegeneric, their cost has decreased greatly andtheir use has expanded to many differentindications, so procurement and availabilityhave not been problematic, but resistanceis more common than amongaminoglycosides. Some foods or beveragesand antacids high in content of divalent or

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e113

trivalent cations can reduce fluoroquinoloneabsorption (16, 167–169). When feasible,taking fluoroquinolones on an emptystomach minimizes the potential of delaysin or reduction of fluoroquinoloneabsorption. Moxifloxacin and to a lesserextent levofloxacin prolong the QT interval.Moxifloxacin may necessitate ECGmonitoring, especially if patients have abaseline QTc. 500 milliseconds or takeother QT-prolonging drugs.

Pharmacokinetic modeling studies forlevofloxacin and moxifloxacin use inchildren are ongoing: recent data suggestthat for children, a levofloxacin dose of atleast 15 to 20 mg/kg daily (170, 171), and amoxifloxacin dose of 10 to 15 mg/kg/d areeffective (based on data showing too lowexposure at 7.5–10 mg/kg) (172). Althougha modeling paper suggests a dose of 25mg/kg/d for infants to 3 months of age and20 mg/kg/d for toddlers, safety at thesedoses has not been verified (173).

Conclusions. In our PS-matchedIPDMA, patients treated with moxifloxacinand levofloxacin had better outcomesthan patients not treated with anyfluoroquinolones, or compared withpatients treated with ofloxacin, afteradjustment for numerous covariates that inthemselves were strong determinants ofoutcome, such as clinical characteristics,extent of drug resistance, and number ofother effective drugs in the treatmentregimen. Moxifloxacin or levofloxacinshould be included in a regimen to achieve atotal of five effective drugs for the treatmentof patients with MDR-TB. Ourrecommendation for the use of moxifloxacinor levofloxacin is strong despite very lowcertainty in the evidence, because we viewedthe significant reduction in mortality,improved treatment success, and relativelyfew adverse effects associated with MDR-TBtreatment that includes these later-generation fluoroquinolones (comparedwith no fluoroquinolones) as having aparticularly favorable balance of benefitsover harms.

Research needs. Research is needed onthe most effective doses (e.g., The OPTI-Q[Prospective, Randomized, Blinded Phase 2Pharmacokinetic/Pharmacodynamic Studyof the Efficacy and Tolerability of IncreasingDoses of Levofloxacin in Combination withOptimized Background Regimen for theTreatment of MDR-TB (TBTC Study 32/DMID Study 13-0057)] Trial;ClinicalTrials.gov Identifier:

NCT01918397) and dosing regimens, aspreliminary work suggests these drugs maybe more effective at higher doses (174). Theactivity, safety, and tolerability of usinghigher doses of fluoroquinolones againstisolates with modestly elevated MICsshould also be explored. Modest increasesin toxicity can be minimized and managedby clinicians and may still be preferable totreating MDR-TB without quinolones.Higher doses may also limit acquiredresistance (175, 176).

Injectables: Amikacin, Capreomycin,Kanamycin, and StreptomycinThe term “injectable drugs” in generalencompasses four drugs: theaminoglycoside antibiotics streptomycin,amikacin, and kanamycin; and the cyclicpolypeptide antibiotic capreomycin (155,162, 177). Aminoglycosides are highlycationic and water soluble, but insoluble inorganic solvents and hydrophobicenvironments, explaining much about theirpharmacology, limited ability to cross lipidmembranes, poor absorption from thegastrointestinal tract, and poor penetrationinto the CNS. Consequently, they areadministered parenterally through slowintravenous infusion or throughintramuscular injection (177).Streptomycin’s core ring structure differs

from all other aminoglycosides, explainingin part why cross-resistance betweenstreptomycin and other aminoglycosides isuncommon (155, 177). These drugs blockprotein synthesis at the ribosomal level bybinding to a highly conserved nucleotidesequence in the prokaryotic 16S ribosomalsubunit, the mRNA decoding region, wheretranslation of mRNA codon to aminoacyl–transfer RNA anticodon normally takesplace (155, 177). Aminoglycosides sharethree important characteristics: 1)concentration-dependent killing, 2) apostantibiotic effect, and 3) synergism withother antibacterial drugs (155). These drugskill bacteria in proportion to drugconcentration, so a single daily dose is moreeffective than divided doses or a continuousinfusion. Moreover, antibacterial activitycontinues many hours after serum levelsbecome undetectable, the so-called“postantibiotic” effect (155).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B. Inour PS-matched IPDMA, treatmentoutcomes for 613 patients who did notreceive an injectable drug were comparedwith 1,554 patients treated with streptomycin,4,330 treated with kanamycin, 2,275 treated

PICO Question 12—Fluoroquinolones: In patients with MDR-TB, are outcomes safelyimproved when regimens include fluoroquinolones compared with regimens that donot include fluoroquinolones?

Recommendation 12: We recommend including moxifloxacin or levofloxacin in aregimen for treatment of patients with MDR-TB (strong recommendation, lowcertainty in the evidence). Our recommendation for the use of moxifloxacin orlevofloxacin is strong despite very low certainty in the evidence because we viewed thesignificant reduction in mortality, improved treatment success, and relatively fewadverse effects associated with MDR-TB treatment that includes these later-generation fluoroquinolones (compared with no fluoroquinolones) as having aparticularly favorable balance of benefits over harms.

PICO Question 13—Injectables: In patients with MDR-TB, are outcomes safelyimproved when regimens include an injectable compared with regimens that do notinclude an injectable?

Recommendation 13: We suggest including amikacin or streptomycin in a regimenfor treatment of patients with MDR-TB when susceptibility to these drugs isconfirmed (conditional recommendation, very low certainty in the evidence). Becauseof their toxicity, these drugs should be used when more effective or less toxic therapiescannot otherwise be assembled to achieve a total of five effective drugs. We suggestNOT including kanamycin or capreomycin (conditional recommendation, very lowcertainty in the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e114 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

with amikacin, and 2,401 treated withcapreomycin. Patients who received two ormore injectable drugs (n=875) wereexcluded, because the outcome could not beascribed to one of the drugs (3). The no-injectable comparison group had notablyfewer AFB smear–positive patients andpatients with cavitary disease. On the otherhand, patients in this group had more pasttreatment with second-line drugs; moreresistance to quinolones, injectables, andXDR TB; and much more treatment withlinezolid. In these respects, the capreomycingroup was similar (more past treatment withsecond-line drugs, worse pretreatment drugresistance) to the other three drugs. Analysiswas based on patients whose isolates weresusceptible to the drug they received, withexceptions noted below.

Benefits. INJECTABLE DRUGS COMPARED

WITH NO INJECTABLE DRUG. Compared with 613patients treated without an injectable drug,1,554 streptomycin-treated patients hadincreased treatment success (aOR, 1.5; 95%CI, 1.1–2.1). In the subgroup withquinolone resistance (17.8% of the total),streptomycin-treated patients had increasedtreatment success (aOR, 3.0; 95% CI,1.3–6.6). Compared with 613 patients nottreated with injectable drugs, 2,275amikacin-treated patients had increasedtreatment success (aOR, 2.0; 95% CI,1.5–2.6). In the subgroup with quinoloneresistance, amikacin was also associatedwith increased treatment success (aOR, 3.0;1.6–5.6). Among patients with XDR-TB,amikacin was associated with reduceddeaths (aOR, 0.4; 95% CI, 0.2–0.8). Incontrast, neither kanamycin norcapreomycin was associated with anybenefit on treatment success or death versusno injectable drug at all. To the contrary,kanamycin treatment was associated withfewer treatment successes (aOR, 0.5; 95%CI, 0.4–0.6). Capreomycin was associatedwith an increased risk of death (aOR, 1.4;95% CI, 1.1–1.7). In XDR-TB, capreomycinwas associated with increased deaths (aOR,3.4; 95% CI, 2.7–4.3) when compared withregimens with no injectable drug.

In a pediatric IPDMA, the use ofsecond-line injectable agents (amikacin,kanamycin, capreomycin) in childrenwith confirmed MDR-TB was associatedwith more treatment success compared withthose not receiving second-line injectableagents (aOR, 2.94; 95% CI, 1.05–8.28;P= 0.041) (113). However, a highproportion of children with less-severe

disease who received no second-lineinjectable agents still did well; therefore,children may be able to be spared frominjectables and the associated toxicities ifnewer, more-effective drugs can beincluded in an all-oral regimen (113).

INJECTABLE DRUGS COMPARED AGAINST

EACH OTHER. Compared with streptomycin-treated patients, patients treated withamikacin had increased treatment successes(aOR, 1.7; 95% CI, 1.3–2.2) but with nosignificant impact on death (aOR, 1.0; 95%CI, 0.8–1.2). Kanamycin and capreomycinwere both significantly inferior tostreptomycin in every respect. Amikacinwas superior to kanamycin and capreomycinin every respect, with higher treatmentsuccess rates, lower death rates, or both. Inthe quinolone-resistant subgroup, on theother hand, use of amikacin did not have asignificant effect on treatment success (aOR,1.7; 95% CI, 0.9–3.4) or death (aOR, 1.2;95% CI, 0.7–2.0) compared with use ofstreptomycin.

Harms. All aminoglycosides andcapreomycin share important toxicities,especially nephrotoxicity, ototoxicity, andelectrolyte disturbances, as well as other lesscommon toxicities (155, 162, 177). Theototoxicity can be vestibular, resulting inloss of balance, or cochlear, resulting inhearing loss. Because these drugs date backto the early years of antibiotic discoveryand development, there is vast publishedexperience with their toxicities. In thetreatment of MDR-TB, the risk of toxicity issubstantial, because the duration oftreatment is many months. Ototoxicity maybe severe and irreversible but with closemonitoring can be minimized or prevented.Nephrotoxicity is often reversible whenidentified early and addressedappropriately; however, ototoxicity canprogress even after the drug is stopped.Monitoring (measuring drug levels;monthly high-quality audiometry,electrolytes, and serum creatinine) andskilled management can prevent or mitigatethese effects and are part of standardpractice for MDR-TB experts. Risk ofhearing loss increases with increasingduration of treatment. For aminoglycosidesin general, the estimated frequency ofnephrotoxicity is 5% to 15% and ofototoxicity 2% to 14%, including 2% to 10%cochlear and 3% to 14% vestibular (155). Ina survey of clinical trials performedbetween 1975 and 1982 and totalingz10,000 patients, the incidence of amikacin

nephrotoxicity was 8.7% (178). Cumulativedose and duration predicted toxicity; olderage, dehydration/hypovolemia/hypotension,prior aminoglycoside treatment, coexistinghepatic or renal disease, and concomitantmedications were important as well (155).Significant hearing impairment exceeds50% in some reported series, renaldysfunction approaches 50%, andvestibular dysfunction as high as 20%has been reported (155, 162, 177).At the other extreme, series have beenpublished with no significant or permanenttoxicity, including with longer-term use(155, 162, 177). Direct comparisonsof drug toxicity for the four drugs usedin TB are scarce. Experience suggestsstreptomycin may be more ototoxic,but that may be a consequence of farlonger and wider use of streptomycin.Observational studies suggest amikacin maybe more ototoxic than the other drugs(179–181).

In children, ototoxicity (hearing loss)has been documented in up to 24% of casesin a retrospective study, which also bringsserious implications for developmentof normal speech (182). Pain ofintramuscular injections can be safelyreduced by adding lidocaine to amikacininjections without interference ofpharmacokinetics (183). Dose range ofamikacin should be 15 to 20 mg/kg assingle daily dose.

Additional considerations. Wheninjectables are used, serum creatinine,electrolyte measurements, clinicalassessment for vertigo and tinnitus, high-quality audiometry (including hearingfrequency of 6,000–8,000 Hz, as high-frequency hearing loss is seen initially), andclinical examinations should be conductedat least monthly, or more frequently ifadverse effects occur. Limited data suggestthere may be a genetic predisposition forhearing loss associated with specificmitochondrial gene mutations (184).Although routine genetic testing is notcurrently suggested, the provider should beaware of these genetic mutations and therisks of ototoxicity (185, 186). Becausebactericidal activity is concentrationdependent, high peak levels and single dailydosing are preferred to divided doses.Because of the postantibiotic effect, toxicitycan be minimized by allowing troughlevels between doses to remain belowdetectable levels for many hours. Someexperts take advantage of these

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e115

pharmacokinetic and pharmacodynamicproperties with thrice-weekly dosing,especially after conversion of cultures tonegative, decreasing healthcare systemdemands and the requirement fordaily injections (16, 177). Intramuscularinjections are painful, and 6 months ofinjections can be traumatic, especiallyto younger patients. Patients’ values maydiffer sharply from providers in thisrespect and should be considered whendetermining whether to use injectableagents.

Conclusions. In our PS-matchedIPDMA, the use of amikacin andstreptomycin when the patient’s isolate wassusceptible to these drugs was associatedwith an increase in treatment success whencompared with the control group notreceiving these injectables. However,because of their toxicity and modestefficacy compared with other drugs, whichalso are less toxic, these drugs should bereserved for when more-effective or less-toxic therapies cannot be assembled toachieve a total of five effective drugs. In ouranalyses, amikacin and streptomycin hadsimilar aORs for treatment success in aminority of patients with fluoroquinoloneresistance. Kanamycin and capreomycinwere ineffective. We recommend againstusing kanamycin or capreomycin. As isthe case for adults, the use of amikacinand streptomycin for children shouldalso be reserved to when more-effectiveor less-toxic therapies cannot beassembled to achieve a total of fiveeffective drugs.

Research needs. N-acetylcysteine, athiol-containing antioxidant, may limit theseverity and irreversibility ofaminoglycoside-induced ototoxicity (187,188), warranting additional research intothis and other otoprotective measures thatmay improve the balance of benefits andharms for using these injectable agents.

LinezolidLinezolid is an oxazolidinone antibioticthat inhibits bacterial protein synthesis bypreventing the fusion of 30S and 50Sribosomal subunits (189). It also binds tohuman mitochondria and inhibits proteinsynthesis, which is the mechanism oftoxicity in clinical use (190, 191).Linezolid was initially used off-label in theabsence of consistent scientific evidence aspart of the regimen for difficult-to-treatcases of MDR- and XDR-TB (192). Thefirst large retrospective observationalstudy suggested linezolid was effective, butwith frequent and often severe adverseevents (192). The same study suggested,for the first time, that reducing the dailydose from 1,200 mg to 600 mg per daymight be associated with fewer adverseevents and improved tolerability. Severalsystematic reviews, IPDMAs, and onecontrolled clinical trial have beenpublished on linezolid for treatment ofMDR-TB (87, 193–196). The clinical trialconfirmed previous observational findingsand, in particular, the effectiveness oflinezolid and its potential toxicity (195). Ameta-analysis with 121 cases of patientswith MDR-TB from 11 countries, treatedwith linezolid, confirmed linezolideffectiveness (culture conversion, 93.5%;treatment success, 81.8%) and that the 600-mg daily dose was safer than the 1,200-mgdose (46.7% adverse events vs. 74.5%,respectively) without lowering itseffectiveness (196). The clinical trialconfirmed the efficacy, safety, andtolerability of linezolid in patients withXDR-TB (194). There are insufficient dataregarding the effectiveness of initiatingtreatment with doses ,600 mg daily torecommend lower doses. Recently, theimportance of therapeutic drugmonitoring to reduce adverse eventspotentially due to linezolid has beenemphasized (197, 198).

Summary of the evidence. Linezolidwas used in regimens for MDR- and XDR-TB across 38 studies (3). The initial dose oflinezolid was 1,200 mg/d for 91 patients infive studies, 600 mg/d for 784 patients in 28studies, and 300 mg/d for 99 patients in fivestudies (3). Procedures and methodology toassemble and rank the certainty in theevidence are reported in APPENDIX A, withevidence profiles for PICO questionsreported in APPENDIX B.

Benefits. In our PS-matched IPDMA,patients who received linezolid-containingregimens were more likely to achievetreatment success (aOR, 3.4; 95% CI,2.6–4.5) and to have a lower rate of death(aOR, 0.3; 95% CI, 0.2–0.3) than those whodid not receive linezolid. The effect ontreatment success was more pronouncedwhen studies from only high-incomecountries were included (aOR, 3.9; 95% CI,2.6–5.8). The greatest impact was found inpatients with XDR-TB, where the aOR forsuccessful treatment versus failure orrelapse was 6.3 (95% CI, 3.9–10.1) and theaOR for death was 0.1 (95% CI, 0.1–0.2).When PS-matched pairs analyses wereperformed comparing the effects ofbedaquiline and linezolid with those of nobedaquiline or linezolid, the combinationof bedaquiline and linezolid was associatedwith an aOR of 2.7 (95% CI, 1.5–4.9) forsuccess versus failure/relapse, and of 0.3(95% CI, 0.2–0.4) for death versussuccess/failure/relapse. The efficacy oflinezolid in children with TB has beenshown in two studies of children ,18 yearsof age, albeit with few patients (199, 200).

Harms. Adverse effects associated withlinezolid in patients with TB includeneurotoxicity (i.e., peripheral neuropathyand optic neuritis), myelosuppression,hyperlactatemia, and diarrhea, all of whichare presumably secondary to the inhibitionof mitochondrial protein synthesis (190,201). A published systematic review of 12studies conducted in 11 countries globallyreported an adverse event rate of 58.9%(hematological, neurological, andgastrointestinal), predominantly noted inindividuals treated with a dosages .600mg/d (196). Hematological toxicity canoccur quickly after starting treatment andcan involve any cell line. Neurotoxicity,including optic neuritis and peripheralneuropathy, occur later, usually after 12 to20 weeks of treatment. Toxicity has beenassociated with trough levels of .2.0

PICO Question 14—Linezolid: In patients with MDR-TB, are outcomes safely improvedwhen regimens include linezolid compared with regimens that do not includelinezolid?

Recommendation 14: We suggest including linezolid in a regimen for the treatmentof patients with MDR-TB (conditional recommendation, very low certainty in theevidence). This is a conditional recommendation despite linezolid-containingregimens showing large reduction in mortality and improved treatment success,similar to bedaquiline and later-generation fluoroquinolones, because linezolid hadmore adverse effects and the balance of benefits and harms was less favorablecompared with those drugs.

AMERICAN THORACIC SOCIETY DOCUMENTS

e116 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

(see ROLE OF THERAPEUTIC DRUG

MONITORING IN TREATMENT OF MDR-TB)(202–204). Adverse effects, especiallymyelosuppression, are also noted to becommon at the currently recommendeddose of 10 mg/kg twice daily in children,10 years of age.

Additional considerations. Althoughuse of linezolid for the FDA-approved 28days or less for non-TB indications isassociated with an acceptable adverse effectprofile (205), the data on the longer-termuse necessary for MDR-TB are limited (3,195, 200). Strict clinical monitoring forpotential toxicity (in particular peripheralneuropathy, optic neuritis, anemia, andleukopenia) is necessary because of the riskof adverse events associated with the long-term use of linezolid (16). If optic neuritisoccurs, many patients may be rechallengedsuccessfully with linezolid once visionnormalizes. Assessment for visual toxicitymust continue after restarting linezolid.Some patients are able to be rechallengedwith the full dose; others are able to avoidrecurrent visual toxicity with a reduceddose of linezolid at 300 mg daily (195).Linezolid should generally not beadministered to patients takingserotonergic agents, such as monoamineoxidase inhibitors, because of the potentialfor serious CNS reactions, such as serotoninsyndrome. Because monoamine oxidasetype A deaminates serotonin, and selectiveserotonin reuptake inhibitors potentiate the

action of serotonin by inhibiting itsneuronal reuptake, administration oflinezolid concurrently with a selectiveserotonin reuptake inhibitor can lead toserious reactions, such as serotoninsyndrome or neuroleptic malignantsyndrome–like reactions (16). Onerandomized clinical trial demonstrated thatlowering the dose from 600 mg/d to 300mg/d after culture conversion reducedtoxicity (195). For children, one modelingstudy reported that linezolid dose of 15mg/kg in full-term neonates and infantsaged <3 months and 10 mg/kg in toddlers,administered once daily, achievedcumulative fraction of response of >90%,with ,10% achieving linezolid areaunder the concentration-versus-time curve(AUC) of 0 to 24 associated with toxicity(173). On the basis of modeling ofpharmacokinetic data from 48 children,WHO and Sentinel Project recommendpediatric doses of linezolid as 15 mg/kgonce daily for children ,15 kg and 10 to12 mg/kg once daily for thoseweighing .15 kg (7, 17). It is commonpractice for patients on linezolid to beprescribed vitamin B6 (16).

Conclusions. In our PS-matchedIPDMA, patients who received linezolid-containing regimens were more likely toachieve treatment success and to have alower rate of death than those who did notreceive linezolid. We suggest includinglinezolid in a regimen for the treatment of

patients with MDR-TB. This is aconditional recommendation despitelinezolid-containing regimens showinglarge reduction in mortality andimproved treatment success, similar tobedaquiline and later-generationfluoroquinolones, because linezolidhad more adverse effects and thebalance of benefits and harms wasless favorable compared with those drugs.

Research needs. Clinical trials ofcombinations of new chemical entities pluslinezolid, including when administered atdifferent doses and durations to optimize itstherapeutic effect while minimizing toxicityare underway (Nix-TB, ClinicalTrials.govIdentifier: NCT02333799; ZeNix [Safetyand Efficacy of Various Doses andTreatment Durations of Linezolid PlusBedaquiline and Pretomanid in Participantswith Pulmonary TB, XDR-TB, Pre-XDR-TB or Non-responsive/Intolerant MDR-TB], NCT03086486). Further linezolidpharmacokinetic and safety data are neededto find optimal dosing in adults andchildren.

Macrolides: Azithromycin andClarithromycinThe macrolides azithromycin andclarithromycin have unclear efficacy androle in the treatment of MDR-TB (23).Macrolides are commonly used to treatupper and lower respiratory tract infectionsand have an essential role in the treatmentof nontuberculous mycobacteria(206). They are believed to haveimmunomodulatory and antiinflammatoryeffects. M. tuberculosis has intrinsic,inducible resistance to clarithromycin (207,208), and in vivo murine TB modelsconfirm the lack of activity of macrolides(209). Clarithromycin may increaselinezolid serum exposure when used incombination (210), prompting someconsideration of potential synergy betweenmacrolides and other MDR-TB drugs (211).WHO does not recommend use ofmacrolides to treat MDR-TB (23).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles for PICOquestions reported in APPENDIX B. In our PS-matched IPDMA, patients who receivedmacrolides (n=1,067) were more likely tohave been treated with second-line drugs andto have resistance to fluoroquinolones or any

PICO Question 15—Macrolides: In patients with MDR-TB, are outcomes safelyimproved when regimens include macrolides compared with regimens that do notinclude macrolides?

Recommendation 15: We recommend NOT including the macrolides azithromycinand clarithromycin in a treatment regimen for patients with MDR-TB (strongrecommendation, very low certainty in the evidence). Our recommendation againstthe use of the macrolides azithromycin and clarithromycin in MDR-TB treatment isstrong despite the evidence being judged to be of very low certainty because we viewedthe increased mortality and decreased likelihood of treatment success associated withthe use of this drug class as having a notably unfavorable balance of benefits to potentialharms.

PICO Question 16—p-Aminosalicylic Acid: In patients with MDR-TB, are outcomessafely improved when regimens include p-aminosalicylic acid compared withregimens that do not include p-aminosalicylic acid?

Recommendation 16: We suggest NOT including p-aminosalicylic acid in atreatment regimen for patients with MDR-TB (conditional recommendation, very lowcertainty in the evidence). When the individualized treatment regimen for patientswith MDR-TB contains newer-generation, more-effective drugs, the addition ofp-aminosalicylic acid does not appear to provide a benefit.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e117

second-line injectable (3). Patients whoreceived macrolides were more likely toreceive later-generation fluoroquinolones,capreomycin, and linezolid. In adjustedanalyses, patients who received macrolideswere less likely to achieve treatmentsuccess (aOR, 0.6; 95% CI, 0.5–0.8) and had ahigher rate of death (aOR, 1.6, 95% CI,1.2–2.0) than patients who did not receivemacrolides.

Benefits. The available evidence doesnot support the use of macrolides in thetreatment of MDR-TB.

Harms. In our PS-matched IPDMA,patients who received macrolides wereless likely to achieve treatment success(aOR, 0.6; 95% CI, 0.5–0.8) and had ahigher rate of death (aOR, 1.6; 95% CI,1.2–2.0) than patients who did not receivemacrolides.

Conclusions. Macrolides, specificallyazithromycin or clarithromycin, shouldnot be included in a regimen for thetreatment of patients with MDR-TB.Our recommendation against the use ofthe macrolides azithromycin andclarithromycin in MDR-TB treatment isstrong despite the evidence being judged tobe of very low certainty because we viewedthe increased mortality and decreasedlikelihood of treatment success associatedwith the use of this drug class as having anotably unfavorable balance of benefits topotential harms.

Research needs. Further research maybe warranted on newer-generationmacrolides and to elucidate if there ispotential synergy of macrolides withlinezolid or other second-line agents.

p-Aminosalicylic AcidOne of the first agents found to beeffective against TB (212), p-aminosalicylicacid has been widely used clinically,although its precise mode of actionremains uncertain (213). With thediscovery of other more potent drugs,including rifampin, p-aminosalicylic acid,initially used combined with streptomycin,was no longer considered in first-lineregimens. It is now used as part of atreatment regimen for MDR- and XDR-TB,although the benefits of p-aminosalicylicacid are not clear and toxicity limits itsuse. Current guidance recommendsthat p-aminosalicylic acid be used tocompose a regimen when five effectivedrugs cannot otherwise be assembled(16, 23).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B. Inour PS-matched IPDMA, p-aminosalicylicacid was not associated with any benefit ontreatment success (aOR, 0.8; 95% CI,0.7–1.0) and was associated with increaseddeath (aOR, 1.2; 95% CI, 1.1–1.4) (3). Thepotential to cause adverse events (12.2% ina previous meta-analysis) has also limitedthe tolerability of p-aminosalicylic acid(23).

Benefits. No association with anybenefit on treatment success was identifiedfor p-aminosalicylic acid in our PS-matchedIPDMA. In an IPDMA of children withMDR-TB, there was no benefit identifiedfor the inclusion of p-aminosalicylic acid inregimens (aOR, 0.75; 95% CI, 0.25–1.96;P= 0.483) (113).

Harms. Gastrointestinal distress iscommon with p-aminosalicylic acid butreported to occur less with the PASERformulation than with older preparations(16). Rare hepatotoxicity andthrombocytopenia have been reported,as has reversible hypothyroidism,particularly when used concomitantlywith ethionamide (120). In the setting ofhypothyroidism, some experts providethyroid replacement therapy rather thandiscontinuing p-aminosalicylic acid.Clinical experience has shown p-aminosalicylic acid to be better toleratedregarding gastrointestinal disturbance inchildren than in adults, althoughhypothyroidism also remains common inchildren.

Additional considerations. Indirectcomparison suggests that ethionamidemay be better than p-aminosalicylic acidif a fifth drug is needed to construct aregimen with at least five effective drugs.Old studies have shown p-aminosalicylicacid to be a good companion drug toprotect other drugs from developingresistance and also have shown efficacy incombination with first-line drugs (214,215). It has recently been suggested bypediatric DR-TB experts that inchildren p-aminosalicylic acid mayreplace an injectable agent in the absence ofbetter drugs (83). When p-aminosalicylicacid is used, experts recommendmonitoring thyroid-stimulating hormone,electrolytes, blood counts, and liverfunction tests.

Conclusions. When the individualizedtreatment regimen for patients withMDR-TB contains newer-generation,more-effective drugs, the addition ofp-aminosalicylic acid does not appear toprovide benefit.

Research needs. Given the limitedarmament of TB drugs available fortreating MDR-TB, some experts haveadvocated for the evaluation of dose-optimized p-aminosalicylic acid toimprove efficacy while minimizingtoxicity (216). Research on whetherp-aminosalicylic acid, among other agents,may offer protection against theacquisition of resistance to other drugs inthe regimen would also be valuable.

PyrazinamidePyrazinamide, a nicotinamide analog, is aprodrug that is converted in vivo intopyrazinoic acid, which interferes withmycobacterial fatty acid synthase.Pyrazinamide has demonstratedeffectiveness against M. tuberculosis; it isincluded in standard regimens fortreatment of drug-susceptible TB and alsoused in regimens for MDR-TB (11, 16, 23).However, recent population-based studiesconducted as part of multicountrysurveillance activities have shown thatpyrazinamide resistance is highly associatedwith rifampin resistance (217, 218). Thisfinding, in conjunction with evidence thatpyrazinamide efficacy is reduced in thesetting of pncA gene mutations (219–222),underscores the importance ofdocumenting drug susceptibility topyrazinamide by WGS, molecular tests, ortraditional DST, if the drug is to beincluded as part of a regimen for MDR-TB.

PICO Question 17—Pyrazinamide: Inpatients with MDR-TB, are outcomessafely improved when regimensinclude pyrazinamide compared withregimens that do not includepyrazinamide?

Recommendation 17: We suggestincluding pyrazinamide in a treatmentregimen for patients with MDR-TB,when the M. tuberculosis isolate hasnot been found to be resistant topyrazinamide (conditionalrecommendation, very low certainty ofevidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e118 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Table 9. Doses of Drugs for Treatment of Adults and Children with Multidrug-Resistant Tuberculosis

Step DrugRoute of

Administration Adults ChildrenReduced Renal

Function*

1 Levofloxacin p.o./i.v. 750–1,000 mg daily† 15–20 mg/kg/d once daily† 3 times/wkMoxifloxacin p.o./i.v. 400 to (600–800 mg)

daily‡10–15 mg/kg/d once daily No change needed

2 Bedaquiline p.o. 400 mg daily3 14 d then200 mg 3 times/wk

>12 yr and >30 kg: adultdose

No change needed

Studies ongoing in lower agegroups and weights. Basedon expert opinion, forchildren .6 yr and weight15–30 kg, half the adultdose can be used (200mg/d32 wk then 100 mgM/W/F for 22–24 wk)

Linezolid p.o./i.v. 600 mg daily >12 yr: 10 mg/kg oncedaily (300 or 600 mg)

No change needed

,12 yr: based on modeledpharmacokinetic data forlower weight bands (7,17, 310):

5–9 kg: 15 mg/kg oncedaily

10–23 kg: 12 mg/kg oncedaily

.23 kg: 10 mg/kg oncedaily

3 Clofazimine p.o. 100 mg daily 2–5 mg/kg/d No change neededCycloserine/terizidone p.o. 250–750 mg dailyx to

achieve serum 20–35mg/L in plasmajj

15–20 mg/kg/d Start with 250 mg daily andverify with TDM in settingof renal disease

4 Amikacin i.v./i.m. 15 mg/kg daily. Someclinicians prefer 25mg/kg 3 times/wk

15–20 mg/kg/d Patients with decreasedrenal function mayrequire the 15-mg/kgdose to be given only 2–3times/wk to allow for drugclearance

Streptomycin i.v./i.m. 15 mg/kg daily. Someclinicians prefer 25mg/kg 3 times/wk

15–20 mg/kg daily or 25–30mg/kg twice weekly¶

Patients with decreasedrenal function mayrequire the 15-mg/kgdose to be given only 2–3times/wk to allow for drugclearance

5 Delamanid p.o. 100 mg twice daily >35 kg: adult dose Mild to moderate renalinsufficiency: no change.Severe insufficiency:limited data, use withcaution

>6 yr and 20–34 kg: 50 mgtwice daily

.3–5 yr and 10–20 kg: 25 mgtwice daily

Lower age/weight: studiesongoing

Ethambutol p.o./i.v. Low dose (companiondrug): 15 mg/kg daily

20–25 mg/kg/d 3 times/wk

High dose (bacteriostaticdrug): 25 mg/kg daily

Pyrazinamide p.o. 25–40 mg/kg daily 30–40 mg/kg/d 3 times/wk

(Continued )

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e119

Pyrazinamide has been demonstrated tohave substantial sterilizing activity as partof combination regimens and has allowedfor treatment shortening in drug-susceptible TB (223, 224). Higher doseshave been shown to be more effective inanimal models and phase 2A studies, butdoses of 40 to 70 mg/kg were found to betoo toxic to be pursued further in humanstudies (224).

Summary of the evidence. Proceduresand methodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B.Our PS-matched IPDMA comparedoutcomes in 1,986 individuals with isolates

susceptible to pyrazinamide who receivedthe drug with 307 individuals with isolatessusceptible to pyrazinamide who did notreceive the drug (3). However, the groupswere not balanced with regard to othereffective drugs; in particular, the patientsnot receiving pyrazinamide were morelikely to receive linezolid and a later-generation fluoroquinolone. Thus, theability to detect an effect of pyrazinamidemight have been partially obscured. In aprevious IPDMA, before linezolid andbedaquiline were more commonly used,administration of pyrazinamide amongpatients with susceptible isolates wasassociated with an aOR of 1.9 (95% CI,1.3–2.9) for cure/complete versus

failure/relapse and an aOR of 1.6 (95% CI,1.3–2.1) for cure/complete versusfailure/relapse/death, compared withpatients receiving pyrazinamide whoseisolates were resistant in vitro (6).

Benefits. In our PS-matched IPDMA,treatment success was significantly lesslikely with regimens containingpyrazinamide (aOR, 0.7; 95% CI, 0.5–0.9),but death was also significantly lessfrequent (aOR, 0.7; 95% CI, 0.6–0.8). Thisparadox may be due to confounding in ourPS-matched IPDMA, as patients who didnot receive pyrazinamide were substantiallymore likely to receive linezolid. Moreover,our PS-matched IPDMA did not assess thepotential ability of pyrazinamide to

Table 9. (Continued )

Step DrugRoute of

Administration Adults ChildrenReduced Renal

Function*

6 Ethionamide p.o. 15–20 mg/kg total (usually250–500 mg once ortwice daily)**

15–20 mg/kg total (divided1–2 times/d)

No change needed

Prothionamide p.o. 15–20 mg/kg total (usually250–500 mg once ortwice daily)**

15–20 mg/kg total (divided1–2 times/d)

No change needed

Imipenem–cilastatin i.v. 1,000 mg 3–4 times/d (imipenem component)15–25 mg/kg/dose 4times/d

May reduce frequency

Meropenem i.v. 1,000 mg 3 times/d†† 20–40 mg/kg/dose 3 times/d May reduce frequencyClavulanate (componentof amoxicillin–clavulanate) forcoadministration withcarbapenems(imipenem–cilastatinand meropenem)

p.o./i.v. 250 mg 3 times/d 25 mg/kg/dose of amoxicillincomponent 3 times/d

May reduce frequency tomatch carbapenem

p-Aminosalicylic acid p.o./i.v. 4 g 2–3 times/d‡‡ 200–300 mg/kg/d in twodivided dosesxx

No change needed

High-dose isoniazidjjjj p.o./i.v. 15 mg/kg daily 15–20 mg/kg/d No change needed

Definition of abbreviations: M/W/F=Monday/Wednesday/Friday; TDM= therapeutic drug monitoring.Updated and modified from References 11, 16, 343, and 344.*Dosages may not apply to patients with severely decreased kidney function, including in the setting of dialysis, for which consultation with a nephrologistis advised.†Levofloxacin doses of up to 1,250 mg have been used safely when needed to achieve therapeutic concentrations. A recent population pharmacokineticstudy in South African children found that higher levofloxacin doses from 18 mg/kg/d for younger children, up to 40 mg/kg/d for older children, may berequired to achieve adult-equivalent exposures (170).‡Higher moxifloxacin doses have been used safely when the isolate is resistant to ofloxacin and the minimum inhibitory concentration for levofloxacin ormoxifloxacin suggests higher doses may overcome resistance. Higher doses also are used in cases of malabsorption.xCycloserine doses can be divided if needed (typically twice daily). Doses .750 mg are difficult for many patients to tolerate.jjCycloserine dose may be lowered if serum concentrations exceed 35 mg/ml, even if patient is not experiencing toxicity, to prevent central nervous systemtoxicity.¶Modified from adult intermittent dose of 25 mg/kg, and accounting for larger total body water content and faster clearance of injectable drugs in mostchildren. Dosing can be guided by serum concentrations.**Ethionamide can be given at bedtime or with a meal to reduce nausea. Experienced clinicians suggest starting with 250 mg once daily and graduallyincreasing the dose over 1 week. Serum concentrations may be useful in determining the appropriate dose. Few patients tolerate 500 mg twice daily.††Studies are ongoing evaluating meropenem at higher doses (ClinicalTrials.gov identifiers: NCT03174184 and NCT02349841).‡‡Some experts prescribe p-aminosalicylic acid at 6 g, and up to12 g, administered once daily (16, 216).xxFor children, some experts prescribe p-aminosalicylic acid at 200 mg/kg administered once daily (216).jjjjIsoniazid is tested at two concentrations. Some experts use these results (or resistance conferred through mutations in inhA) to select a higher dosewhen it tests resistant at the lower concentration and susceptible at the higher concentration. The higher dose may achieve in vivo concentrationssufficiently high to overcome low-level resistance (16, 216).

AMERICAN THORACIC SOCIETY DOCUMENTS

e120 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

contribute to treatment shortening, as hasbeen achieved in drug-susceptible TB.Because pyrazinamide is associated withincreased success only when used to treatpatients whose isolates are susceptible tothe drug (144), whenever feasible, thedecision to include pyrazinamide in aregimen should be based on pyrazinamidesusceptibility results.

In an IPDMA of children with MDR-TB, the addition of pyrazinamide to aregimen showed no benefit in the treatmentof confirmed MDR-TB cases (aOR, 1.63;95% CI, 0.41–6.56; P=0.484) (113).Pyrazinamide resistance was, however, nottested and improved selection of cases on thebasis of resistance might change this outcome.

Harms. Pyrazinamide has been usedextensively in the treatment of TB, andtoxicities are well documented (225). Themost common is gastrointestinal upset orintolerance. In a recent review of thetolerability of TB drugs, pyrazinamide wasassociated with serious adverse events in 56of 2,023 (2.8%) patients (226). This isconsistent with previous reports ofpyrazinamide toxicity (11, 225, 227).Hepatic enzyme elevations are commonwith pyrazinamide, and significanthepatotoxicity, although less common, canoccur. Modest elevated serum uric acidlevels are also expected, although theclinical significance of this is unclear.Nongouty polyarthralgias andhypersensitivity reactions can occur. Flaresof clinical gout can also occur, especially inthose with a history of prior gouty arthritis.

Additional considerations. All patientsreceiving pyrazinamide as part of an MDR-TB treatment regimen should be monitoredcarefully for signs or symptoms ofhepatotoxicity and have their pyrazinamidedose held or decreased if such toxicity isdetected. Isolated increases in uric acidwithout symptoms of gout are common andare not an indication to discontinue thedrug. Recent population-based studies havefound that pyrazinamide resistance iscommon in the setting of MDR with someregional variability, suggesting thatpyrazinamide susceptibility should beconfirmed or suspected if the drug isincluded in the regimen (217, 228, 229).Although there are known challengesrelated to accurately determiningphenotypic DST for pyrazinamide, recenthighly predictive DNA sequencingtechniques show significant promisefor newer genomic approaches (230).

When included in the regimen, mostexperts use doses of 25 to 40 mg/kg/dorally. The recommended dose ofpyrazinamide in children is 30 to 40mg/kg daily.

Conclusions. We suggest includingpyrazinamide in a regimen for the treatmentof patients with MDR-TB, when the M.tuberculosis isolate has not been foundresistant to pyrazinamide.

Research needs. Pyrazinamide is beingevaluated as part of novel treatmentregimens for both DS and MDR-TB inmultiple clinical trials (56). Development ofa reliable, simple molecular test forpyrazinamide susceptibility is a criticallyimportant research need. Doseoptimization studies for pyrazinamide arewarranted, as higher doses may be moreefficacious but also may be more toxic.

Building a TreatmentRegimen for MDR-TB

The guideline committee proposes a clinicalstrategy tool for building a treatmentregimen for MDR-TB (Table 10). Theclinical strategy tool incorporates theevidence-based review of the individualdrugs, with consideration of the balance ofbenefits and harms for each drug, theexperience of MDR-TB experts on thecommittee, as well as perspectives ofpatients. This clinical strategy toolencourages the building of all oral regimenswith five effective drugs (to which theisolate is susceptible or has low likelihoodof resistance) for the treatment of MDR-TB. In our PS-matched IPDMA, significantfavorable synergies were identified withimproved treatment success and reducedmortality when bedaquiline was used incombination with linezolid or clofazimine.As noted, amikacin and streptomycin showmodest effectiveness when the patient’sisolate is susceptible to these drugs;however, because of their significanttoxicities, aminoglycosides should bereserved for when a more-effective or less-toxic regimen cannot otherwise beassembled. The final choice of drugs anddrug classes is contingent on many factors,including patient preferences, harms andbenefits associated with agents, the capacityto appropriately monitor for significantadverse effects, consideration of drug–druginteractions, comorbidities, and drugavailability. Final regimen development,

therefore, is individualized and may differsubstantially from the approach describedin Table 10. Doses of the drugs for treatingadults and children with MDR-TB areprovided in Table 9, modified and updatedfrom the 2016 ATS/CDC/IDSA Treatmentof Drug-Susceptible TB Practice Guidelines(11).

Role of Therapeutic DrugMonitoring in Treatment ofMDR-TB

Specific pharmacokinetic (PK)/pharmacodynamic (PD) targets, especiallythe AUC divided by the MIC over 24 hours(AUC0–24/MIC), are being recognized asplaying an important role in determiningefficacy (231). This has been demonstratedmost clearly in hollow fiber systems in vitrothat isolate the activity of the drug againstthe organism (232, 233). Further evidencehas been provided by animal models(mouse, rabbit, nonhuman primates) andthrough human clinical trials (234–237).Drug levels are usually chosen to be four tofive times greater than the MIC. Because ofthe complexities of clinical disease, it ismore challenging to isolate the PK/PDcontribution of individual drugs, butstudies have nonetheless been informativeon the relationships between efficacyand drug exposure (238–240). In humanTB, a combination of drugs is used, andeach patient has his or her uniqueduration of disease, host genetics, andparticular strain of M. tuberculosis. Thegeneral term for these PK/PD data is“exposure–response” data (i.e., for agiven amount of drug exposure, howmuch response can you expect?). Muchof the published data focus on the first-lineTB drugs, with some emerging databecoming available for second-line drugs(241–243).

In clinical practice, the actual MIC foreach drug often is not available.Epidemiological cut-off values or “criticalconcentrations” that separate wild-typefrom more resistant isolates can be usedfor selecting what drugs to include in aregimen (244). These in vitro cut-offs arebased on patterns of susceptibilitycompared with achievable concentrationswithin humans. An organism is not just“susceptible” as an inherent property; it issusceptible to inhibition or killing byspecific, tested concentrations of the

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e121

drugs. Individual MIC values might bepreferred; however, in practice, thereare technical and financial barriers tosuch individualized data. Followingstandardized dosing, clinical experienceover the past 3 decades clearly shows thatsome patients have low drugconcentrations, leading to clinical failures(11, 231, 241, 242, 244).

Many MDR-TB experts use therapeuticdrug monitoring (TDM) to identify patientswith problems with drug absorption,thereby informing dose adjustments.

Individuals who have poor response to TBtreatment despite adherence may benefitfrom TDM (11). Patients with TB withgastrointestinal problems that increase riskof malabsorption, concurrent HIVinfection, impaired renal clearance, ordiabetes should be prioritized for TDM(11). Furthermore, some experts use TDMfor all patients being treated for MDR orXDR-TB early in treatment, rather thanwaiting for a poor response. TDM shouldbe used and interpreted in consultationwith an expert in MDR-TB.

TDM also provides patient-specificinformation that may help limit toxicity dueto certain drugs, including the injectabledrugs cycloserine and linezolid (245–247).In particular, linezolid toxicity is associatedwith elevated trough values (202–204).“Target” ranges for the injectable drugs andcycloserine have been proposed, andresearch continues on refining these.Fluoroquinolones display concentration-dependent efficacy, and higher doses ofmoxifloxacin and levofloxacin arebeing studied (174, 248). Currently,

Table 10. Clinical Strategy to Build an Individualized Treatment Regimen for MDR-TB

d Build a regimen using five or more drugs to which the isolate is susceptible (or has low likelihood of resistance), preferably withdrugs that have not been used to treat the patient previously.

d Choice of drugs is contingent on capacity to appropriately monitor for significant adverse effects, patient comorbidities, andpreferences/values (choices therefore subject to program and patient safety limitations).

d In children with TB disease who are contacts of infectious MDR-TB source cases, the source case’s isolate DST result should be used ifan isolate is not obtained from the child.

d TB expert medical consultation is recommended (ungraded good practice statement).

Step 1: Choose one later-generation fluoroquinolone LevofloxacinMoxifloxacin

Step 2: Choose both of these prioritized drugs BedaquilineLinezolid

Step 3: Choose both of these prioritized drugs ClofazimineCycloserine/terizidone

Step 4: If a regimen cannot be assembled with five effectiveoral drugs, and the isolate is susceptible, use one of theseinjectable agents*

AmikacinStreptomycin

Step 5: If needed or if oral agents preferred over injectableagents in Step 4, use the following drugs†

Delamanid‡

PyrazinamideEthambutol

Step 6: If limited options and cannot assemble a regimenof five effective drugs, consider use of the following drugs

Ethionamide or prothionamidex

Imipenem–cilastatin/clavulanate or meropenem/clavulanatejj

p-Aminosalicylic acid¶

High-dose isoniazid**

The following drugs are no longer recommended forinclusion in MDR-TB regimens:

Capreomycin and kanamycinAmoxicillin/clavulanate (when used without a carbapenem)Azithromycin and clarithromycin

Definition of abbreviations: DST=drug susceptibility testing; INH= isoniazid; IPDMA= individual patient data meta-analyses; MDR=multidrug-resistant;PS=propensity score; TB= tuberculosis.*Amikacin and streptomycin should be used only when the patient’s isolate is susceptible to these drugs. Because of their toxicity, these drugs should bereserved for when more-effective or less-toxic therapies cannot be assembled to achieve a total of five effective drugs.†Patient preferences in terms of the harms and benefits associated with injectables (the use of which is no longer obligatory), the capacity to appropriatelymonitor for significant adverse effects, consideration of drug–drug interactions, and patient comorbidities should be considered in selecting Step 5 agentsover injectables. Ethambutol and pyrazinamide had mixed/marginal performance on outcomes assessed in our PS-matched IPDMA; however, someexperts may prefer these drugs over injectable agents to build a regimen of at least five effective oral drugs. Use pyrazinamide and ethambutol only whenthe isolate is documented as susceptible.‡Data on dosing and safety of delamanid are available in children >3 years of age.xMutations in the inhA region of the Mycobacterium tuberculosis genome can confer resistance to ethionamide/prothionamide as well as to INH. In thissituation, ethionamide/prothionamide may not be a good choice unless the isolate is shown to be susceptible with in vitro testing.jjDivided daily intravenous dosing limits feasibility. Optimal duration of use not defined.¶Fair/poor tolerability and low performance. Adverse effects reported to be less common in children.**Data not reviewed in our PS-matched IPDMA, but high-dose isoniazid can be considered despite low-level isoniazid resistance but not with high-levelINH resistance.

AMERICAN THORACIC SOCIETY DOCUMENTS

e122 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

avoiding low serum concentrations seemsadvisable.

A common clinical practice is to collectsamples at 2 and 6 hours after drugadministration to measure concentrationsthat may distinguish normal drugabsorption (a 2-h value within the normalrange) from delayed absorption (a 6-h valuegreater than the 2-h value, approaching thenormal range) and malabsorption (bothvalues are below the normal range). Thisapproach also works for injectable drugs.Furthermore, trough values for linezolid canbe helpful (11, 231, 241, 242, 244). Although itis clearly possible to cure patients withoutTDM and even in the setting of lower serumdrug concentrations, the available datasuggest that the probability of cure decreaseswith decreasing drug concentrations(231–241). Given the incomplete informationcurrently available, many expert clinicians usein vitro susceptibility data and TDM asavailable tools for optimizing the treatment ofpatients with MDR- and XDR-TB. For allpatients, but in particular those receivingoutpatient treatment, TDM should becoordinated and conducted using patient-centered approaches.

Shorter-Course,Standardized, 9- to 12-MonthRegimen for MDR-TB

A randomized, phase 3, noninferiority trial,STREAM Stage 1 (ClinicalTrials.govidentifier: NCT02409290) was recentlyconducted to assess a shorter-courseregimen composed of existing drugs forMDR-TB (8). This regimen had achievedsuccess rates of 83% (95% CI, 71.0–90.3%)in cohort studies (249, 250). The shorter-course regimen is standardized andcomposed of kanamycin, moxifloxacin (inplace of gatifloxacin, which was thefluoroquinolone originally used in the“Bangladesh” regimen), prothionamide,clofazimine, pyrazinamide, high-doseisoniazid, and ethambutol for an initialperiod (4–6 mo) and moxifloxacin,clofazimine, pyrazinamide, and ethambutolfor the continuation phase (5 mo) (23).The total duration of therapy is between9 and 11 months, compared with anindividualized regimen of 18 to 24 monthsof therapy. The medication costs of theshorter regimen are believed to be less thanthat of conventional regimens, currently

undergoing a formal economic evaluationas part of the STREAM trials (251).

Summary of the EvidenceIn the STREAM Stage 1 clinical trial, of 424participants who underwent randomization,383 were included in the modifiedintention-to-treat population (8). Favorablestatus was reported in 79.8% of participantsin the long-regimen group and in 78.8% ofthose in the short-regimen group—adifference, with adjustment for HIV status,of 1.0 percentage point (95% CI,27.5 to 9.5). The results with respect tononinferiority were consistent among the321 participants in the per-protocolpopulation. Procedures and methodologyto assemble and rank the certainty in theevidence are reported in APPENDIX A, withevidence profiles for PICO questionsreported in APPENDIX B. In our PS-matchedIPDMA of 12,030 patients (n= 50 studies),169 (n= 33 studies) were eligible foranalysis of death and 1,369 (n= 33 studies)were eligible for analysis of treatmentsuccess (249) once the criteria used for theshorter-course regimen were applied (249).There were data from 532 individuals (n= 3studies) on the shorter regimen availablefor analysis of death and 498 (n= 3 studies)for analysis of treatment success. Afterusing PS matching to adjust for age, sex,HIV, smear status, past TB treatment withfirst-line drugs, and number of effectivedrugs, there were no statistically significantassociations for the shorter regimen withtreatment success (aOR, 0.5; 95% CI,0.02–13), or deaths (aOR, 1.7; 95% CI,

0.6–4.6), compared with individualizedregimens.

BenefitsThe shorter regimen allows for asignificantly shorter duration of therapycompared with the individualized regimenand consequently less pill burden,medication cost, and associated provideradministration costs. The burden onpatients of lost productivity or lost wagesand out-of-pocket costs is considerably lesswith a shorter regimen. From preliminaryresults of the STREAM Stage 1 trial, therewas a documented reduction in costs topatients due to less time away from work,fewer clinic visits, and less spending onsupplementary food (252).

HarmsIn the STREAM Stage 1 clinical trial, anadverse event of grade 3 or higher occurredin 45.4% of participants in the long-regimengroup and in 48.2% in the short-regimengroup (8). Prolongation of either the QTinterval or the QTc to 500 millisecondsoccurred in 11.0% of participants in theshort-regimen group, compared with 6.4%in the long-regimen group (P= 0.14). Deathoccurred in 8.5% of participants in theshort-regimen group and in 6.4% in thelong-regimen group, and acquiredresistance to fluoroquinolones oraminoglycosides occurred in 3.3% and2.3%, respectively. Among the participantswho had HIV coinfection at baseline, 18 of103 (17.5%) in the short-regimen groupdied, compared with 4 of 50 (8.0%) in the

PICO Question 18—Shorter-course, standardized regimen: In patients with MDR-TB,does treatment with a standardized MDR-TB regimen for <12 months lead to betteroutcomes than treatment with an MDR-TB regimen for 18–24 months?

Recommendation 18: The shorter-course regimen is standardized with the use ofkanamycin (which the committee recommends against using) and includes drugs forwhich there is documented or high likelihood of resistance (e.g., isoniazid,ethionamide, pyrazinamide). Although the STREAM Stage 1 randomized trial foundthe shorter-course regimen to be noninferior to longer injectable-containing regimenswith respect to the primary efficacy outcome (8), the guideline committee cannotmake a recommendation either for or against this standardized shorter-courseregimen, compared with longer individualized all-oral regimens that can be composedin accordance with the recommendations in this practice guideline. We make aresearch recommendation for the conduct of randomized clinical trials evaluating theefficacy, safety, and tolerability of modified shorter-course regimens that includenewer oral agents, exclude injectables, and include drugs for which susceptibility isdocumented or highly likely.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e123

long-regimen group (HR in a post hocanalysis, 2.23; 95% CI, 0.76–6.60) (8).Similarly, in a recently published IPDMA, agreater proportion of individuals whoreceived the shorter regimen had lesstreatment success and more death (249). Inthe IPDMA, other adverse effects were alsostatistically insignificant with the shorterregimen, including deafness and ototoxicity(relative risk, 1.5; 95% CI, 0.6–4.0), liverinjury (relative risk, 2.2; 95% CI, 0.5–10.3),hepatitis (relative risk, 2.5; 95% CI,0.3–21.2), and renal impairment (relativerisk, 4.5; 95% CI, 0.6–35.2). One of themost concerning adverse effects of theshorter regimen is hearing loss, with 7.1%reported in the African study and between0% and 23% in the meta-analysis (249,250). The STREAM Stage 1 trial found thatear and labyrinth disorders occurred in7.4% participants on the shorter regimen,compared with 5.7% with the longerregimen, which was not statisticallysignificant (8); the frequency may havebeen lower than in cohort studies becausehearing loss was not monitored byaudiometry in STREAM Stage 1.

Additional ConsiderationsWhen applying the eligibility criteria fromWHO for using the shorter-course,standardized regimen to the populationincluded in our PS-matched IPDMA(7, 253), .15% of individuals would havebeen eligible for the standardized, shorterregimen. In Europe, patient eligibility forthe shorter-course regimen has rangedfrom 7.9% (48 of 612 new cases) in a studyperformed at TB reference centers (254) to16.9% in a surveillance-based study

performed by the European Centre forDisease Prevention and Control (ECDC)(93, 107, 255). In the United States, .15%of patients with MDR-TB would be eligiblefor the shorter-course MDR-TB regimen(256). Data from California showed either14.6% or 20.5% eligibility, based onwhether high-dose isoniazid orethionamide would be determined effectiveon the basis of genetic katG or inhAmutations, respectively (257). Theavailability of molecular and growth-basedDST is key to determining the potentialeligibility of shorter-course regimen use.The combined use of both modalities canbetter inform drug resistance and potentialtreatment regimens. High-dose isoniazid islikely to be active against organisms withlow-level isoniazid resistance, commonlyassociated with a mutation in the inhA genethat also confers resistance to ethionamide(257). Ethionamide is more likely to beactive against M. tuberculosis organismswith high-level resistance to isoniazid,associated with a mutation in katG andwhich are less commonly resistant toethionamide (257). Global data have shownthat katG is present in 64.3% of the casestested and inhA in 19.2%, with someregional differences (258). The shorterregimen includes both ethionamide andhigh-dose isoniazid and therefore is likelyto be effective against both of thesecommon MDR-TB resistance patterns (257,259). There is uncertainty regarding thediagnostic accuracy and reproducibility ofethambutol and pyrazinamide growth-based DST assays (260). For ethambutol,the ECDC has reported that thegrowth-based DST when performed on

Mycobacteria Growth Indicator Tube(MGIT) is reliable (107), whereas ModelPerformance Evaluation Program data haveidentified variability in detectingethambutol resistance, with the majority oflaboratories having disagreement on one toseveral strains with ethambutol resistance(261). Alternatively, molecular testing forpncA mutations as a method fordetermining pyrazinamide resistance maybe more reliable, because on average 85% ofTB strains resistant to pyrazinamide willhave such a mutation (217, 221, 222). Inlimited-resourced settings, testing all thedrugs composing the shorter-courseregimen may not be possible (106, 108). Incontrast, in Europe, DST for ethambutol(93, 107) and pyrazinamide (106) is wellstudied, but katG and inhA geneticmutations are not. In the United States,growth-based DST data are available forethambutol, pyrazinamide, high- and low-level isoniazid, and ethionamide, andmolecular testing is available through somestate public health laboratories and throughthe CDC’s MDDR service (257). Therefore,considering patients with either katG orinhA (but not both) mutations to be eligiblefor the shorter regimen might be rational.Last, the rate of cross-resistance betweenofloxacin and moxifloxacin is likely notcomplete. Data from the ECDC showedthat 81% of M. tuberculosis isolates thatwere resistant to ofloxacin were alsoresistant to moxifloxacin. However, othersettings (e.g., Bangladesh and Pakistan)have shown cross-resistance as low as 7%(107).

ConclusionsThe shorter-course regimen was judged by theguidelines committee to have minimaldesirable effects (on treatment success,mortality, and culture conversions) andsmall to moderate undesirable effects(adverse events, limited applicability, andthe use of kanamycin as part of thestandardized regimen), and includes drugsfor which there is documented or highlikelihood of resistance (e.g., isoniazid,ethionamide, and pyrazinamide).Although the WHO’s STREAM Stage 1randomized trial found the shorter-courseregimen to be noninferior to a long regimenwith respect to the primary efficacyoutcome (8), the guideline committeecannot make a recommendationeither for or against this standardizedshorter-course regimen compared

PICO Question 19—Surgery for MDR-TB: Should elective lung resection surgery(i.e., lobectomy or pneumonectomy) be used as an adjunctive therapeutic option incombination with antimicrobial therapy, versus medical therapy alone, for adults withMDR-TB?

Recommendation 19a: We suggest elective partial lung resection (e.g., lobectomy orwedge resection), rather than medical therapy alone, for adults with MDR-TBreceiving antimicrobial-based therapy (conditional recommendation, very lowcertainty in the evidence). The writing committee believes this option would bebeneficial for patients for whom clinical judgement, supported by bacteriological andradiographic data, suggests a strong risk of treatment failure or relapse with medicaltherapy alone.

Recommendation 19b: We suggest medical therapy alone, rather than includingelective total lung resection (pneumonectomy), for adults with MDR-TB receivingantimicrobial therapy (conditional recommendation, very low certainty in theevidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

e124 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

with longer individualized regimens.We instead make a researchrecommendation for the conduct ofrandomized clinical trials evaluatingthe efficacy, safety, and tolerability ofmodified shorter-course regimens thatinclude newer oral agents, excludeinjectables, and include drugs for whichsusceptibility is confirmed or deemed tobe highly likely. If this shorter-courseregimen is used, we recommend obtainingDST for all medications in the regimen,with the exception of clofazimine,for which reliable testing is notavailable, and recommend carefulside effect monitoring, includinghigh-quality audiometry, monthlymicrobiologic monitoring, and closecase management, especially in personswith HIV.

Research NeedsWemake a research recommendation for theconduct of randomized clinical trialsevaluating the efficacy, safety, and tolerabilityof modified shorter-course regimens thatinclude newer oral agents, excludeinjectables, and include drugs for whichsusceptibility is confirmed or deemed tobe highly likely. Further research isneeded on medications such as linezolid,bedaquiline, and other medicationscurrently in clinical trials as substitutionsin the regimen if patients experienceadverse effects or resistance developsto any of the medications in theregimen (262). Research on modifiedshorter-course regimens for pediatricpatients and in individuals living withHIV is also needed. Until more dataregarding the use and outcomes of theshorter-course regimens in patients

with HIV are available, this treatmentapproach should preferentiallybe considered only within a researchstudy. Current trials are underway to helpanswer some these questions (262).

Role of Surgery in MDR-TB

Surgery was one of the first therapeuticapproaches for treating TB. It was replacedby chemotherapy between 1960 and 1975.However, several scientific societies andnational and international organizationssuggest consideration of surgery as anadjunctive therapy for MDR-TB. This isbased on the results of observationalretrospective studies. Selectedindications are highlighted, including failureof drug therapy, relapse, localized (e.g., anisolated cavity) or extensive pulmonaryTB, and clinical complications (e.g.,hemoptysis or empyema) (23, 263–270).

Summary of the EvidenceSystematic reviews and meta-analyses havebeen performed on the role of surgery inpatients with MDR- and XDR-TB (263, 271,272). The main limitation of systematicreviews of surgery in MDR and XDR-TBthat summarize results of observationalstudies combining study-level data is thetremendous variability in patientcharacteristics, background chemotherapyregimens, and types of surgical procedures(263, 271, 272). An IPDMA of surgery inMDR-TB was designed to address thoseshortcomings (271). Procedures andmethodology to assemble and rank thecertainty in the evidence are reported inAPPENDIX A, with evidence profiles forPICO questions reported in APPENDIX B.

The IPDMA identified 67 cohort studies, ofwhich 45 could not be used because eitherindividual patient data were not available orthe surgical status of patients was notknown. Twenty-six studies comprising6,431 patients with MDR-TB were included(271). We used data from this IPDMA togenerate the evidence profiles. Despite theanalytic advantages of an IPDMA, whichallows for adjustment on baselineimbalance in many prognostic factors, therewas a substantial residual risk of bias in theresults. There is no information about theimpact of surgery on adverse effects orquality of life.

BenefitsPatients with partial lung resection had ahigher probability of treatment success, asopposed to treatment failure, relapse, ordeath (aOR, 3.0; 95% CI, 1.5–5.9). However,the estimate is very uncertain because of thelimitations of individual studies and lack ofprecision in results.

HarmsIn the published IPDMA, a substantiallyhigher proportion of patients who had apneumonectomy died (8.5%) comparedwith those who had a partial resection(2.2%), but the authors could not establishwhether patients died as a result of surgicalcomplications or of their TB (271). Theestimates of the effects of pneumonectomyon risk of death (aOR, 1.8; 95% CI, 0.6–5.1)and treatment success (aOR, 0.8; 95% CI,0.1–6.0) were not statistically significant. Asfor partial lung resection, both estimates arevery uncertain. Treatment success inpatients with XDR-TB was noted to belower when patients underwent surgerycompared with patients who did not (aOR,

PICO Question 20—Treatment of isoniazid-resistant TB: PICO Question 20a: Should patients with isoniazid-resistant TB be treatedwith a regimen composed of a fluoroquinolone, rifampin, ethambutol, and pyrazinamide for 6 months compared with rifampin,ethambutol, and pyrazinamide (without a fluoroquinolone) for 6 months?

PICO Question 20b: Should patients with isoniazid-resistant TB be treated with a regimen composed of fluoroquinolone, rifampin,and ethambutol for 6 months and pyrazinamide for the first 2 months compared with a regimen composed of a fluoroquinolone,rifampin, ethambutol, and pyrazinamide for 6 months?

Recommendation 20a: We suggest adding a later-generation fluoroquinolone to a 6-month regimen of daily rifampin, ethambutol,and pyrazinamide for patients with isoniazid-resistant TB (conditional recommendation, very low certainty in the evidence).

Recommendation 20b: In patients with isoniazid-resistant TB treated with a daily regimen of a later-generation fluoroquinolone,rifampin, ethambutol, and pyrazinamide, we suggest that the duration of pyrazinamide can be shortened to 2 months in selectedsituations (i.e., noncavitary and lower-burden disease or toxicity from pyrazinamide) (conditional recommendation, very lowcertainty in the evidence).

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e125

0.4; 95% CI, 0.2–0.9), an effect that washeterogeneous across studies and mayalso be confounded by factors thatpredisposed to poor outcomes in thesepatients (271). Alternatively, patients whowere healthy enough to withstand surgerymay have intrinsically lower mortality, so itis difficult to predict the direction ofpotential bias.

Additional ConsiderationsIn the MDR-TB treatment guidelines updatedin 2016, WHO recommended elective partialsurgery as an intervention complementary tochemotherapy in specifically selectedindividuals (23). The committeeacknowledged there is significant uncertainty,with a paucity of publications and dataregarding patient preferences, acceptability bysurgical personnel, cost, and feasibility ofperforming elective surgery as an interventionfor MDR-TB.

ConclusionsOn the basis of the limited evidence that isavailable, there appears to be a net benefitfrom an elective partial lung resection (e.g.,lobectomy or wedge resection) when offeredtogether with a recommended MDR-TBregimen, compared with medical therapyalone. Committee members believed thatthis therapeutic option would probably bemore beneficial when clinical judgement,supported by bacteriological and radiographicdata, suggests a strong risk of relapse ortreatment failure with medical regimen alone.We found no currently available evidence thatpneumonectomy would be beneficial forpatients with MDR TB receiving abackground drug regimen.

Research NeedsRigorously designed and conducted studies,ideally well-done randomized trials thatmeasure and report benefits but also adverseeffects and quality of life, are needed to clarifythe role of surgery in the management ofpatients with MDR-TB. The following specificissues need to be addressed: the optimaltiming for surgery, the optimal drug regimensand their duration before and after surgery,the role of surgery in special populations andpatients with comorbid conditions (e.g., thoseliving with HIV), the optimal surgicalapproaches, the optimal infection controlmeasures to be implemented perioperatively,and the role of pulmonary rehabilitation (263,267, 270, 273).

Treatment of Isoniazid-Resistant TB

Isoniazid is an important first-line agent forthe treatment of TB, possessing potent earlybactericidal activity against M. tuberculosis.However, monoresistance to isoniazid isfrequent worldwide, with an estimatedprevalence of 8% (range, 5–11%) of TB cases(12). A recent systematic review and meta-analysis compared the treatment outcomes ofisoniazid-resistant TB to outcomes of drug-susceptible TB and found that treatment ofisoniazid-resistant TB with first-line drugsresulted in suboptimal outcomes, with highertreatment failure (11% vs. 1%) and relapse(10% vs. 5%) (274). In addition, the studyfound that standardized empirical treatmentof new isoniazid-resistant TB cases may becontributing to higher rates of acquired drugresistance (8% vs. 0.3%). Prior ATS, CDC,and IDSA guidelines recommendedtreatment with a standard four-drug regimen(isoniazid, rifampin, pyrazinamide, andethambutol) for 6 months, withdiscontinuation of isoniazid after the resultsof DST are known and if isoniazid resistancewas found (275).

Summary of the EvidenceProcedures and methodology to assembleand rank the certainty in the evidence arereported in APPENDIX A, with evidenceprofiles for PICO questions reported inAPPENDIX B. An IPDMA of 33 datasets with6,424 patients, of whom 3,923 patients in 23studies received regimens related toisoniazid-resistant TB, was used as theevidence (276). Regimens of interest for ouranalyses (all with or without isoniazid) were1) rifampin, ethambutol, and pyrazinamide;and 2) rifampin, ethambutol, andpyrazinamide, plus a fluoroquinolone. Forthese analyses, isoniazid-resistant TB wasdefined based on phenotypic resistance toisoniazid and susceptibility to rifampicin,with or without additional resistance topyrazinamide, ethambutol, or streptomycin.Critical concentrations of 0.1 mg/ml or 0.2mg/ml were used in 21 out of 23 centers forthe definition of resistance to isoniazid(276). For PICO Question 20a, we found fewstudies that used regimens of 6 months ofrifampin, ethambutol, and pyrazinamide,plus a fluoroquinolone, but several thatused regimens of 8 to 9 months of thisregimen; hence, we combined 6 with .6months for this PICO. For the

comparator of rifampin, ethambutol,pyrazinamide with or without isoniazid,we similarly combined regimens of6 and .6 months’ duration regimens,after comparison between these durationsshowed outcomes were not significantlydifferent. For PICO Question 20b, fewpatients received regimens thathad a fluoroquinolone, rifampin, ethambutol,and a shorter duration of pyrazinamide. Atotal of 118 patients received 1 to 3 months ofpyrazinamide in conjunction with 6and .6 months’ durations of rifampin,ethambutol, and fluoroquinolone-containingregimen and were available for analyses.

BenefitsCompared with 6 months of daily rifampin,ethambutol, and pyrazinamide(with or without isoniazid), adding afluoroquinolone to this regimen wasassociated with significantly greater treatmentsuccess (aOR, 2.8; 95% CI, 1.1–7.3) but withno significant effect on mortality (aOR, 0.7;95% CI, 0.4–1.1) or acquired rifampicinresistance (aOR, 0.1; 95% CI, 0.0–1.2). Whenevaluating the impact of shortening theduration of pyrazinamide (ranging from1–3 mo) in a regimen that contains afluoroquinolone, the treatment success wasvery high, with 117 of 118 patients achievingtreatment success. On the other hand,comparisons of shorter pyrazinamideregimens to regimens including both afluoroquinolone and pyrazinamide for >6months did not show significantly differentresults (aOR, 5.2; 95% CI, 0.6–46.7). Similarresults were obtained when the comparisonswere restricted to patients receiving later-generation fluoroquinolones, namelymoxifloxacin, levofloxacin, and gatifloxacin(data not shown).

HarmsThe outcome of adverse events from TBdrugs was intended to be assessed inthe IPDMA but could not be analyzedbecause these outcomes were eithernot reported or reported with verydifferent definitions. The adverse eventsassociated with TB drugs, especiallypyrazinamide, are well established(227, 277).

Additional ConsiderationsWe note that the estimates of effect wereimprecise from the IPDMA because of thesmall number of patients who received the

AMERICAN THORACIC SOCIETY DOCUMENTS

e126 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

regimens of interest. Given thatpyrazinamide is the most toxic of the presentfirst-line drugs, a key potential advantage ofadding a fluoroquinolone would be toshorten the duration of pyrazinamide to the

initial 2 months of treatment. Although theIPDMA only had 118 patients receivingfluoroquinolone-containing regimens withshorter durations of pyrazinamide, it is notedthat treatment success was very high in this

group (117 of 118). On the basis of theefficacy signals seen and the known toxicitiesof prolonged pyrazinamide, the committeeviewed the balance of benefits and harms tofavor shortening the duration of

Table 11. Select Antiretroviral and Non–Rifamycin-based Antituberculosis Drug Overlapping Toxicities and Potential AdverseDrug–Drug Interactions

Potential Overlapping Toxicities andDrug–Drug Interactions Antiretroviral Drugs Non-Rifamycin TB Drugs

Arrhythmias, QT interval prolongation Lopinavir/ritonavir, efavirenz Fluoroquinolones†, bedaquiline, delamanid,clofazimineNote PR interval prolongation* with

atazanavir, lopinavir/ritonavir

Hepatic cytochrome P450 enzyme systemmetabolism

Induce CYP P450 metabolism: efavirenz,etravirine, and nevirapine

Bedaquiline, delamanid

Impede CYP P450 metabolism: proteaseinhibitors, cobicistat

Nephrotoxicity Tenofovir,‡ atazanavir Aminoglycosides, capreomycinIsolated creatinine elevationx: cobicistat,dolutegravir

Mental health changes (depression,psychosis, dizziness, etc.)

Efavirenz, rilpivirine; dolutegravir,elvitegravir, raltegravir

Cycloserine, isoniazid, ethionamide,fluoroquinolones†

Peripheral neuropathy Stavudine, zidovudine Aminoglycosides, capreomycin, linezolid,isoniazid, ethionamide, cycloserine,fluoroquinolones†

Hepatotoxicity Lactic acidosis with hepatic steatosishigher risk with stavudine, zidovudine;protease inhibitors; nevirapine (higherrisk in patients with elevated CD4 cellcounts); less common with efavirenz,etravirine and rilpivirine; maraviroc

Isoniazid, pyrazinamide, ethionamide,p-aminosalicylic acid, clofazimine

Indirect hyperbilirubinemiajj: atazanavir

Skin rash Nevirapine (higher risk in patients withelevated CD4 cell counts), efavirenz,etravirine, rilpivirine. Any proteaseinhibitor (especially those containingsulfonamide moiety: e.g., darunavir);abacavir (hypersensitive reaction a risk inpatient who are HLA-B5701 positive);raltegravir

All TB drugsNote skin pigmentation withclofazimine use

Thioacetazone should be avoided in peoplewith HIV, because of an elevated risk of asevere adverse skin reaction

Dysglycemia Lopinavir/ritonavir, ritonavir, stavudine,zidovudine

Ethionamide, p-aminosalicylic acid,fluoroquinolones, linezolid

Myelosuppression/cytopenias Zidovudine Linezolid

Lactic acidosis Stavudine, zidovudine Linezolid

Definition of abbreviation: TB= tuberculosis.Saquinavir, indinavir, fosamprenavir, tipranavir, and didanosine are older antiretroviral drugs that are rarely used in the United States. They do have many ofthe adverse interactions listed above with select TB drugs, and clinicians considering the use of these agents should therefore consult with an expert.*Use with caution in patients with underlying cardiac dysrhythmia.†Fluoroquinolones include ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin.‡Tenofovir alafenamide is a prodrug of tenofovir and U.S. Food and Drug Administration approved in 2015. It is associated with decreased incidence ofosteoporosis and nephrotoxicity compared with tenofovir disoproxil fumarate through achieving higher intracellular drug concentrations with a lower doseadministered.xIncreases in serum creatinine via decrease in renal tubular creatinine excretion are commonly seen with cobicistat and dolutegravir usage. This is not atoxicity.jjIndirect hyperbilirubinemia is expected with atazanavir and indinavir. This is not a toxicity.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e127

pyrazinamide when a later-generationfluoroquinolone is included in the regimenin patients in whom there is toxicityanticipated or experienced because ofpyrazinamide or when the patient hasnoncavitary, lower burden of disease. Finally,plasma peak concentration and exposure tomoxifloxacin have been shown to decrease byapproximately 30% when coadministered withrifampin (278, 279). The impact of thesedecreased exposures on outcomes has notbeen established. However, some experts optto use levofloxacin when combined withrifampin.

ConclusionsWe conclude that in patients with isoniazid-resistant TB, the addition of a later-generationfluoroquinolone to 6 months of dailyrifampin, ethambutol, and pyrazinamideimproves treatment success rates. In patientsin whom toxicity from pyrazinamide isanticipated or experienced, or in patients withactive TB with lower burden of disease(i.e., noncavitary), the committee viewed thebalance of benefits and harms to favorshortening the duration of pyrazinamidewhen a later-generation fluoroquinolone isincluded in the regimen, acknowledging thatthe certainty in the evidence is very low andmore research is needed.

Research NeedsAll but 2 of the 23 studies included in theIPDMA were observational. Moreover, theanalyzed population included only 37children, 119 patients with diabetes, and249 patients with HIV infection. Given theburden of isoniazid-resistant TB worldwide,clinical trials that include these specialpopulations and evaluate new regimens,including the evaluation of the efficacy,safety, and tolerability of shorter versuslonger durations of pyrazinamide, are urgentlyneeded.

Treatment of MDR-TB inSpecial Situations

HIV InfectionAmong patients with HIV and TB, rifampinresistance has been identified more oftencompared with those without HIV (280, 281).Patients with MDR-TB and HIV have up to afourfold higher risk of mortality comparedwith patients with MDR-TB without HIV(282). Low CD4 cell counts (e.g., <50cells/ml) in patients with HIV and MDR-TBfurther correlate with higher mortality (283,284). Numerous practice guidelinesrecommend HIV testing of people withsuspected or confirmed TB (11, 20, 34, 285),regardless of drug resistance. Because oftheir high risk for mortality early in thecourse of TB disease, people with HIV inwhom TB is suspected are recommended toreceive rapid testing for TB using nucleic acidamplification tests coupled with moleculardiagnostic DST for rifampin (with or withoutisoniazid) (13, 20).

For patients with HIV receiving therapyfor drug-susceptible TB, studies foundsignificantly lower mortality among patientsreceiving concurrent antiretroviral therapy(ART) compared with those not receivingART (286–288). U.S. practice guidelinesrecommend starting ART within 2 weeks ofinitiating treatment for drug-susceptiblenon-CNS TB for patients with CD4 cellcounts,50 cells/ml and by 8 to 12 weeks forpatients with higher CD4 cell counts (11,289). Although the optimal timing forstarting ART to reduce patient mortality hasnot yet been adequately determined forpatients with MDR-TB, lower mortality rateshave been shown in multiple studies amongpatients with MDR-TB receiving concurrentART compared with those not receivingART (283, 290–297), especially amongpatients with TB with CD4 counts ,50cells/ml (283, 292). The decreased mortality

seen in patients treated for MDR-TB whoconcurrently receive ART, especially amongthose with CD4 cell counts ,50 cells/ml,supports a similar ART managementapproach as recommended for drug-susceptible TB.

MDR-TB of the CNS in patients withHIV presents additional challenges. TBmeningitis with isoniazid-resistant TB, RR-TB, or MDR-TB can be associated withhigher mortality compared with drug-susceptible disease (298–300). Starting ARTearly in patients with TB is associatedwith a higher incidence of immunereconstitution inflammatory syndrome(286, 288, 301), and this can be even moreproblematic in CNS disease. A recent studyin Vietnam of patients with advanced AIDSfound a higher incidence of potentially life-threatening (grade 4) adverse events amongpatients with TB meningitis starting ARTearly compared with those delaying ARTuntil after 2 months of standardized first-line TB treatment and did not show asurvival benefit of starting ART early (302).It has been recommended to delay startingART by 8 weeks in patients with CNS TBand HIV (11); however, there is a paucity ofdata in patients with MDR-TB of the CNS.The optimal approach for initiation of ARTin patients with this medical conditionremains uncertain, and close clinicalmonitoring is warranted.

Drug interactions betweenantiretroviral and anti-TB agents arecommon in the management of patientswith HIV and TB, particularly withrifamycins. Because the rifamycins (with thepossible exception of rifabutin) are not usedfor treatment of MDR-TB, interactions ofother TB medication classes with ARTshould be considered. Bedaquiline and/ordelamanid might be considered for use inpatients with HIV. Although efavirenz canproduce a decrease in serum bedaquilineconcentrations and this combination isavoided, other ART drugs, including theprotease inhibitors and cobicistat, canresult in increased serum bedaquiline levels.The current WHO recommendations forthe use of delamanid apply to patients livingwith HIV (7). Drug–drug interactionstudies in healthy volunteers of delamanidwith tenofovir, efavirenz, andlopinavir/ritonavir show that no doseadjustments were needed (303).Nonetheless, when delamanid is includedin the regimen for MDR-TB in HIV-infected patients, the design of their

PICO Question 21—Treatment of Contacts Exposed to MDR-TB: Should contactsexposed to an infectious patient with MDR-TB be offered LTBI treatment versusfollowed with observation alone?

Recommendation 21: For contacts with presumed MDR LTBI due to exposure to aninfectious patient with MDR-TB, we suggest offering treatment for LTBI (conditionalrecommendation, very low certainty in the evidence). We suggest 6 to 12 months oftreatment with a later-generation fluoroquinolone alone or with a second drug, on thebasis of drug susceptibility of the source-case M. tuberculosis isolate. On the basis ofevidence of increased toxicity, adverse events, and discontinuations, pyrazinamideshould not be routinely used as the second drug.

AMERICAN THORACIC SOCIETY DOCUMENTS

e128 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

ART regimens should be developed inconsultation with HIV and ART experts. Athorough review of all patients’ medicationsshould be performed, in consultation withMDR-TB experts, to select individualizedregimens with less potential foroverlapping ART/TB drug toxicities(Table 11). Useful webpages regardingdrug interactions (TB/HIV and other)are available from AIDSinfo(https://aidsinfo.nih.gov/guidelines), CDC(http://www.cdc.gov/tb/publications/guidelines/TB_HIV_Drugs/default.htm),University of California San Francisco (http://hivinsite.ucsf.edu/insite?page=ar-00-02),University of Liverpool (http://www.hiv-druginteractions.org/), and IndianaUniversity (http://medicine.iupui.edu/clinpharm/ddis/).

The management of MDR-TB is morecomplex among patients with HIVinfection. The higher pill burden ofcombined ART with expanded TB drugtherapy, potential drug–drug interactions,the management of immune reconstitutioninflammatory syndrome, and otherconcurrent HIV-associated opportunisticdiseases all pose unique challenges in thecare of these patients. The management ofpatients with HIV and MDR-TB can bestbe performed by a multidisciplinary careteam composed of health providersexperienced in MDR-TB, HIV, andpublic health case management (304, 305).MDR-TB experts can be found throughCDC-sponsored TB Centers of Excellencefor Training, Education, and MedicalConsultation (http://www.cdc.gov/tb/education/rtmc/default.htm), national,state and local TB control programs(https://www.cdc.gov/tb/links/tboffices.htm),and international MDR-TB expert groupssuch as the Global TB Network (6).

ChildrenOn the basis of recent modeling studies, it isestimated that there are about 1 millionincident cases of TB in children annuallyand 230,000 deaths caused by the disease(306, 307). About 35,000 cases of MDR- andXDR-TB occur in children annually (308,309). Our PS-matched IPDMA did notinclude sufficient numbers of children toallow the formulations of GRADE-basedrecommendations. Nonetheless, on the basisof a recent IPDMA of 975 children withMDR-TB from 18 countries, recentpharmacokinetic studies in children, andseveral observational studies showing good

outcomes, the recommendations noted onchoice of drugs, composition of regimens,and durations of treatment for adults canalso be applied to children with MDR-TB(113, 119, 124, 161, 165, 170, 200, 310).

There are some special considerations forformulating effective regimens for children ofvarious ages and adolescents. The bacterialburden in young children with TB is muchsmaller than that in most adults with TB. As aresult, most drug resistance in children waspresent when the organism was inhaled(primary resistance), and further developmentof resistance while on therapy (secondaryresistance) is much less common in children.However, the paucibacillary nature ofchildhood TB also makes microbiologicconfirmation much more difficult. The onlyway to determine drug susceptibility in casesthat meet clinical definitions of TB disease(i.e., microbiologic confirmation is notavailable) is by linking the child to a specificsource case for whom the drug susceptibilitiesof the organism are known. Linking the childand a specific case is more feasible in low-burden settings to which these guidelines applybut can be difficult in high-burden settingswhen there may be more than one possiblesource case. Also, standard definitions ofrelapse and treatment failure for pediatric TBtrials are inconsistent because the low burdenof organisms in children makes microbiologicconfirmation of these outcomes difficult.

There are several technical factors thatcan affect the outcome of treatment forMDR- and XDR-TB in children. The twoage extremes of childhood have beensomewhat neglected in studies of MDR- andXDR-TB. Little is known about thepharmacokinetics, safety, and tolerability ofthe drugs used to treat drug-resistant TB inneonates, infants, and toddlers (115).Children, especially those ,2 years of age,are more prone to developing disseminatedTB, including meningitis. Drugs thatpenetrate well into the CSF, such aslinezolid, might have an advantage overdrugs that appear to have less penetration,such as ethambutol and bedaquiline.Adolescents can develop TB similar to thatfound either in adults or in young children.However, adolescent patients often havebeen excluded from TB treatment trials.Most of the oral drugs used to treat MDR-and XDR-TB are not licensed for children.Although the Global Drug Facility haspediatric dispersible tablets available(http://stoptb.org/gdf/drugsupply/drugs_available.asp), these formulations are not

registered with FDA or the EuropeanMedicines Agency, which has limitedcommercial availability of child-friendlydosage forms. As a result, for youngerchildren, the medication often has to becrushed or put into suspension or capsulesonce opened, and the pharmacokinetics andpharmacodynamics of these variouspreparations are unknown. HIV-infectedchildren may be exposed to lowerconcentrations of certain orally administereddrugs than HIV-uninfected children giventhe same bodyweight dose. Unfortunately,the pharmacokinetics in HIV-infectedchildren of drugs used to treat MDR- andXDR-TB are largely unstudied. Childrengenerally have a more difficult timetolerating injectable medications because ofpain and the fact that many children withTB are malnourished and have diminishedmuscle mass. With the recent developmentof new oral drugs, it is hoped that injectabledrugs can be avoided in children wheneverpossible. Fortunately, children generallytolerate the oral TB drugs better than adults,with fewer serious adverse events resulting infewer breaks in therapy. Also, most childrenwith TB have not developed the commonchronic diseases of adulthood and will notsuffer complications of them duringtreatment. However, drug adverse effects canbe difficult to assess in children and likelyare underreported. In general, the sameschedules used to monitor adverse eventsand laboratory abnormalities in adultpatients treated for MDR- or XDR-TB alsoshould be used for children.

Outcomes of MDR-TB treatment inchildren. A recently published systematicreview (33 studies) and IPDMA (28 of thestudies) described treatment outcomes for975 children with MDR-TB using randomeffects multivariate logistical regressionadjusted for age, sex, HIV infection,malnutrition, severe extrapulmonarydisease, and severe pulmonary disease onchest radiograph (113). Overall, 78% had asuccessful treatment outcome, including75% of the microbiologically confirmedcases. However, treatment was successful inonly 56% of HIV-infected children who didnot also receive ART during TB treatmentcompared with an 82% success rate in thosealso treated with ART. In children withconfirmed MDR-TB, the use of injectableagents and high-dose isoniazid wasassociated with treatment success.Unfortunately, limitations of this studyincluded that the vast majority of patients

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e129

came from one site (Cape Town, SouthAfrica), difficulty in estimating thetreatment effects of individual drugs withinmultidrug regimens, the availability of onlyobservational cohort studies, and thattreatment decisions were based on theclinician’s perception of illness, withresulting potential for bias.

Conclusions. Excellent treatmentoutcomes have been demonstrated in bothtrials and extensive clinical experience forchildren with MDR- and XDR-TB usingindividualized treatment regimens with thecurrently available drugs. Microbiologiccure and probable cure rates in children canreach 80% to 90% with early recognition ofdrug resistance and adequate treatment(311). The greatest difficulties have beenrecognizing that the child has MDR-TB andthe ability of the child to tolerate injectabledrugs; fortunately, with expandedknowledge of and experience with thenewer oral drugs, such as bedaquiline anddelamanid, pediatricians with expertise inMDR-TB believe that the majority ofchildren with MDR-TB likely can be curedwith an all-oral drug regimen.

Pregnant WomenUntreated MDR-TB during pregnancy canbe associated with adverse maternal andfetal outcomes. Crucial gaps exist in theliterature on treatment of MDR-TB inpregnant women, including theeffectiveness, safety, and tolerability ofavailable treatment regimens, as well astiming and duration of second-line drugs. Insupport of these guidelines, we conducted asystematic literature review with a focus onpregnant women with MDR-TB andincluded all original research reportingMDR-TB treatment outcomes duringpregnancy, including case reports and caseseries, in all languages. We excluded animalstudies, review articles, letters to the editor,and articles without documentation ofMDR-TB treatment during pregnancy. Full-text review and data extraction wereconducted by three reviewers. Our initialsearch yielded 280 publications, of which 16met inclusion criteria for full-text review(312–327); however, 3 publications wereeventually excluded because of lack of dataon medications or outcomes (315, 325,327). The remaining 13 articles wereobservational case reviews without anycomparison groups. Treatment regimensreported were individualized according todrug susceptibility and tolerability of drugs.

Summary of the evidence. Of the 65pregnant women for whom MDR-TBtreatment outcome data were available, 49%(n= 32) were cured and 20% (n= 13)completed treatment for a treatmentsuccess proportion of 69%. Fourteenpercent (n= 9) of the women died.Treatment failure was reported in 9%(n= 6), and 3% (n= 2) were lost to follow-up. Across these studies, four women werestill receiving treatment at the time ofpublication. Fetal outcomes included 78.5%(n= 51) healthy births, with eightchildren being born premature or havinglow birth weight. Medical abortions wereobtained by 12% of the patients (n= 8), and3% (n= 2) underwent spontaneousabortions. Stillbirth was reported for onechild (1.5%), 3% (n= 2) of children wereborn with HIV, and 1.5% (n= 1) hadTB/HIV coinfection.

Conclusions. On the basis of the limiteddata available, we conclude that there isevidence to support treatment of MDR-TBduring pregnancy, including the prescriptionof second-line drugs. Most of the second-linedrugs are pregnancy category C per the U.S.Food and Drug Administration, with theexception of bedaquiline and meropenem,which are classified as category B (accordingto the previous FDA letter-basedclassification system, currently undergoingrevision [328]), and aminoglycosides, whichare category D (71, 320, 329). Despite lowcure rates reported in the literature, webelieve that the benefits of treatment tomother, child, and the community outweighthe harms. There is no evidence to supportone particular regimen; however, mostMDR-TB experts avoid aminoglycosides andethionamide in pregnant women ifalternative agents can be used for an effectivetreatment regimen.

Research needs. Global registries aimedat collecting data on the efficacy, safety, andtolerability, and maternal and fetal outcomesassociated with MDR-TB regimens inpregnant women are urgently needed.Furthermore, given the significant morbidityand mortality associated with MDR-TB,a reconsideration of the current approachof assumed exclusion of pregnantwomen from MDR-TB clinical trials iswarranted. A recent consensus statementfrom an international expert paneladvocated for allowing pregnantand lactating women to remaineligible for phase III MDR-TB trials,

unless there is a compelling reason forexclusion (330).

Treatment of ContactsExposed to MDR-TB

In 1992 and 2000, ATS and CDC advised: 1)no treatment for MDR LTBI for personsnot at high risk for progression to TBdisease, but provision of clinical follow-upfor TB disease signs and symptoms; 2) 6 to12 months of treatment with two or moremedications to which the isolate of thesource case is susceptible (331, 332).

Summary of the EvidenceProcedures and methodology to assembleand rank the certainty in the evidence arereported in APPENDIX A, with evidence profilesfor PICO questions reported in APPENDIX B. Asystematic review of 21 publishedobservational studies that examinedoutcomes of TB incidence, treatmentcompletion, adverse effects, and costeffectiveness was used as the evidence (333).Six articles compared TB incidence amongcontacts receiving MDR LTBI treatmentversus untreated contacts: 10 presented TBincidence only for contacts who receivedMDR LTBI treatment, and 5 presented TBincidence only for untreated contacts.

BenefitsUsing data from five non–registry-matchedcomparison studies included in thesystematic review of 21 publishedobservational studies, MDR-TB incidenceoccurred in 2 of 190 (1.1%) patients treatedfor MDR LTBI, compared with 18 of 126(14.3%) in those who received no MDRLTBI treatment (290). The estimated MDR-TB incidence reduction was 90% (9–99%),using a negative binomial model controllingfor person time and overdispersion (333).

HarmsFrom 11 studies having data by regimen ontreatment discontinuation due to adverseeffects, there was high (51%) treatmentdiscontinuation among patients takingpyrazinamide-containing regimens (333).About one-third of patients takingfluoroquinolone-containing regimens,without pyrazinamide, had adverse effects,but only 2% discontinued treatment. Forchildren <15 years of age regardless ofregimen, treatment discontinuation due toadverse effects was substantially less (5% vs.33%) (333).

AMERICAN THORACIC SOCIETY DOCUMENTS

e130 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Additional ConsiderationsModeled cost-effectiveness studies haveshown greatest benefit using afluoroquinolone-based MDR LTBItreatment. In one study, the most cost-effective regimen wasfluoroquinolone/ethambutol, followed byfluoroquinolone alone, then bypyrazinamide/ethambutol (333). Apyrazinamide/fluoroquinolone regimen wasparticularly toxic, as measured by treatmentdiscontinuation, and prevented about halfas many TB cases as the most cost-effectiveoption (333). An expert panel generated apolicy brief on this topic recently,acknowledged that further evidence isurgently needed, but still endorsed theimmediate implementation of postexposuremanagement of household contacts ofMDR-TB, endorsing the approach as beingeffective, feasible, and cost efficient (334).The Curry International TB Center DrugResistant TB: Clinician’s Survival Guidesuggests an LTBI regimen of levofloxacin ormoxifloxacin alone or combined with asecond medication to which the isolate ofthe infectious source patient is susceptible(16).

Among those with presumed MDRLTBI, there is probable effectiveness ofpreventing TB through MDR LTBItreatment. From systematic reviews ofstudies reporting incidence for both treatedand untreated contacts, TB incidence wassignificantly lower with MDR LTBItreatment (333, 335–339). In studies thatpublished outcomes by regimen, there werehigh treatment discontinuation rates due toadverse effects and toxicity in personstaking pyrazinamide-containing MDR-LTBI regimens (336, 337, 339–342). Therewere low treatment discontinuation ratesdue to adverse effects and toxicity inpersons taking fluoroquinolone-containingMDR-LTBI regimens. Finally, children areat high risk for developing MDR-TB ifinfected with MDR LTBI and generallyexperience fewer adverse effects than adults.Given the uncertainties above, many TBexperts and programs provide extendedpost-treatment completion follow-up forpatients treated for presumed MDR LTBI.

ConclusionsFor contacts with presumed MDR LTBI dueto exposure to an infectious patient withMDR-TB, we suggest offering treatment forLTBI versus following with observation

alone. For treatment of MDR LTBI, wesuggest 6 to 12 months’ treatment with afluoroquinolone alone or with a seconddrug, on the basis of source-case isolateDST. On the basis of evidence of increasedtoxicity, adverse events, anddiscontinuations, pyrazinamide should notbe routinely used as the second drug. In lieuof fluoroquinolone-based treatment, thereare few data for the use of other second-line medications and, because of toxicity,they are not recommended by experts. Forcontacts to fluoroquinolone-resistant,pre–XDR-TB, pyrazinamide/ethambutolmay be an effective option, if source-caseisolate DST shows susceptibility to thesedrugs. In children, TB drugs are generallybetter tolerated, and levofloxacin ispreferred because of the availability of anoral suspension formulation.

Research NeedsRandomized clinical trials on the effectivenessof safer, shorter, more-effective MDR LTBIregimens are underway. Three randomizedclinical trials of MDR LTBI treatment forcontacts to persons with infectious MDR-TBinclude: 1) TB-CHAMP (Tuberculosis ChildMultidrug-Resistant Preventive TherapyTrial): 24 weeks of levofloxacin versusplacebo in children ,5 years of age, 2) VQUIN (Levofloxacin versus Placebo for theTreatment of Latent Tuberculosis amongContacts of Patients with Multidrug-Resistant Tuberculosis [The VQUIN MDRTrial]): 6 months of daily levofloxacin versusplacebo, and 3) PHOENIx (ProtectingHouseholds on Exposure to NewlyDiagnosed Index Multidrug-ResistantTuberculosis Patients [PHOENIx MDR-TB]): 6 months of daily delamanid versus 6months of isoniazid. In addition, moreinformation is needed on the costeffectiveness of treating contacts withpresumed DR-TB infection using thesenewer MDR LTBI regimens.

Summary of Key Differencesbetween ATS/CDC/ERS/IDSAand WHO 2019 ConsolidatedGuidelines on Drug-ResistantTuberculosis Treatment

The 2019WHOConsolidated Guidelines onDrug-Resistant Tuberculosis Treatmentguidelines were based on a modified set ofdata that expanded on the initial individualpatient dataset used for these

ATS/CDC/ERS/IDSA guidelines (3). Forthe 2019 WHO update, data from 625patients were included by WHO fromeight other datasets received in 2018(Australia, Belarus, Brazil, France, Latvia,Republic of Korea, Russian Federation,and the EndTB project), as well as asample of 3,626 patients receivingtreatment in South Africa, including 1,210cases started on bedaquiline in 2015. Withthese newer data added, the WHOindividual patient dataset removed 3,367records because of incomplete DSTdocumentation (7). Despite the changesin the datasets described, WHO andATS/CDC/ERS/IDSA recommendationsare largely concordant, as they werederived concurrently, using a similarapproach (GRADE methodology and amultidisciplinary Guideline DevelopmentGroup), and were informed by anindividual patient data meta-analysis thatoverlapped substantially. The maindifferences are highlighted below.

d WHO recommendations are stated asapplying to MDR and RR-TB. OurATS/CDC/ERS/IDSA guidelines do notaddress management of rifampinresistance in the absence of isoniazidresistance.

d WHO recommendations only retain theintensive/continuation phasedenomination for regimens thatincorporate amikacin and streptomycin.We anticipate that use of regimens thatincorporate injectables will declineover time, and it is likely that bothWHO and ATS/CDC/ERS/IDSArecommendations on treatmentduration will change in the near futureand the distinction between an initialand continuation phase will bereduced further in the context ofnewer, more potent all-oral regimensemerging.

d The duration of the injectable phasein the WHO recommendations relatesto its total length (6–7 mo in 2019update) rather than time afterculture conversion as in theseATS/CDC/ERS/IDSA guidelines. Therecommendation on total duration issimilar in length in both guidelines. Forthe WHO guidelines, no separaterecommendations are made for pre-XDRor XDR-TB.

d The minimum number of medicineslikely to be effective at the start of

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e131

treatment is five in the ATS/CDC/ERS/IDSA guidelines and four in theWHO recommendations. The WHOrecommendation was based onanalyses that showed no clear benefitfrom exceeding four effective drugsat the start of treatment. Thedifference between WHO andATS/CDC/ERS/IDSA guidelines maybe due to the greater use ofbedaquiline and other effectivedrugs in the latter updates of theIPDMA used by WHO. TheATS/CDC/ERS/IDSA recommendationchose five medicines on the basis ofbenefits noted in analyses shown inTables 3 and 4 and also to anticipatetoxicities related to MDR-TB drugs,wherein one or more agents wouldlikely need to be held or permanentlydiscontinued within the first 6 monthsof treatment.

d The WHO recommendations forlinezolid and bedaquiline were strong,with the evidence judged to be ofmoderate certainty by the WHOGuideline Development Group, whereasthe ATS/CDC/ERS/IDSA guidelinesjudged the certainty of the IPDMA tobe of lower certainty. The WHOrecommendation on delamanid was

primarily determined by the outcomesof Trial 213 reported in aggregateformat by the study investigator (asubsequent IPDMA of trial datadid not change the strengthof the recommendation or thepositioning of the drug in terms ofpriority for choice in a longerregimen). The ATS/CDC/ERS/IDSAcommittee concurs with theWHO recommendation thatdelamanid can be used fortreatment of MDR-TB.

d With respect to composing a regimen,Table 10 in these guidelines, providing astepwise selection of agents, is differentfrom the WHO guidance, whichproposes three groupings, essentiallycombining steps. On the basis ofjudgements of the respectiveguideline committee groups, the orderingof drugs within groups is minimallydifferent.

d WHO maintains the recommendation onthe standardized 9- to 11-month shorter-course regimen, underspecific conditions, whereas theATS/CDC/ERS/IDSA committee, on thebasis of the fact that the regimen isstandardized with the use of drugs forwhich there is documented or high

likelihood of resistance (e.g., isoniazid,ethionamide, pyrazinamide), could notmake a recommendation for oragainst the shorter regimen comparedwith individualized regimens. Instead,the ATS/CDC/ERS/IDSA committeemakes a research recommendationfor the conduct of randomizedclinical trials evaluating the efficacy,safety, and tolerability of modifiedshorter-course regimens thatinclude newer oral agents, excludeinjectables, and include drugs forwhich susceptibility is confirmed orhighly likely.

d For patients with isoniazid-resistant TB,the WHO recommendations do notprovide for a reduction of duration ofpyrazinamide use under situations such asrecommended by the ATS/CDC/ERS/IDSAguidance.

d Finally, although the WHOrecommends evaluation of drugresistance, their guidance accepts thatglobally empirical regimens willcontinue to be used. TheATS/CDC/ERS/IDSA guidancerequires microbiologicaldata to create a regimen suitable forthe individual patient’s strain oftuberculosis. n

This official clinical practice guideline was prepared by an ad hoc subcommittee of the ATS, CDC, ERS, and IDSA.

Members of the subcommittee are asfollows:

PAYAM NAHID, M.D., M.P.H.1 (Co-Chair)GIOVANNI BATTISTA MIGLIORI, M.D.2

(Co-Chair)GIOVANNI SOTGIU, M.D., PH.D.3x

(Co-Chair)TERENCE CHORBA, M.D., D.SC., M.P.H.4

(Co-Chair)SUNDARI R. MASE, M.D., M.P.H.4* (Co-Chair)BARBARA SEAWORTH, M.D.5 (Co-Chair)GRAHAM H. BOTHAMLEY, M.D., M.A., PH.D.6

JAN L. BROZEK, M.D., PH.D.7‡

ADITHYA CATTAMANCHI, M.D., M.A.S.1

J. PETER CEGIELSKI, M.D., M.P.H.4

LISA CHEN, M.D.1

CHARLES L. DALEY, M.D.8

TRACY L. DALTON, Ph.D.4

RAQUEL DUARTE, M.D., PH.D., M.P.H.9,10

FEDERICA FREGONESE, M.D.11

C. ROBERT HORSBURGH, JR., M.D.12

FAIZ AHMAD KHAN, M.D.C.M., M.P.H.13

FAYEZ KHEIR, M.D., M.S.C.R.14x

ZHIYI LAN, M.SC.13

ALFRED LARDIzABAL, M.D.15

MICHAEL LAUZARDO, M.D.16

JOAN M. MANGAN, PH.D., M.S.T.4

SUZANNE M. MARKS, M.P.H., M.A.4

LINDSAY MCKENNA, M.P.H.17

DICK MENZIES, M.D.13

CAROLE D. MITNICK, SC.D.18

DIANA M. NILSEN, M.D.19

FARAH PARVEZ, M.D., M.P.H.4,19

CHARLES A. PELOQUIN, PHARM.D.16

ANN RAFTERY R.N., P.H.N., M.S.1

H. SIMON SCHAAF, M.D.20

NEHA S. SHAH, M.D., M.P.H.4jj

JEFFREY R. STARKE, M.D.21

JOHN W. WILSON, M.D.22

JONATHAN M. WORTHAM, M.D.4

*Present address: World Health Organization,Southeast Asian Regional Office, New Delhi,India.‡Lead methodologist.xMethodologist.jjPresent address: Division of AIDS, NationalInstitute of Allergy and Infectious Diseases,National Institutes of Health, Bethesda,Maryland.

1University of California San Francisco, SanFrancisco, California; 2Istituti Clinici ScientificiMaugeri IRCCS, Tradate, Italy; 3University ofSassari, Sassari, Italy; 4Centers for DiseaseControl and Prevention, Atlanta, Georgia;5University of Texas Health Science Center,Tyler, Texas; 6Queen Mary University of Londonand London School of Hygiene and TropicalMedicine, London, United Kingdom; 7McMasterUniversity, Hamilton, Ontario, Canada; 8NationalJewish Health, Denver, Colorado; 9Instituto deSaude Publica, Porto, Portugal; 10CentroHospitalar de Vila Nova de Gaia/Espinho, VilaNova de Gaia, Portugal; 11Universite deMontreal, Montreal, Quebec, Canada; 12BostonUniversity, Boston, Massachusetts; 13McGillUniversity, Montreal, Quebec, Canada;14Tulane University Health Sciences Center,New Orleans, Louisiana; 15New JerseyMedical School, Newark, New Jersey;16University of Florida, Gainesville, Florida;17Treatment Action Group, New York, New York;18Harvard Medical School, Boston,Massachusetts; 19New York City Department ofHealth and Mental Hygiene, New York, NewYork; 20Stellenbosch University, Cape Town,South Africa; 21Baylor College of Medicine,Houston, Texas; and 22Mayo Clinic, Rochester,Minnesota

AMERICAN THORACIC SOCIETY DOCUMENTS

e132 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

Author Disclosures: G.H.B. served as aspeaker for Johnson & Johnson. C.L.D. servedon an advisory committee for Insmed, OtsukaPharmaceutical, Horizon, Johnson & Johnson,Paraek, and Spero; served on a data and safetymonitoring board for Otsuka Pharmaceutical andSanofi; and received research support fromInsmed. C.R.H. served on a data and safetymonitoring board for Janssen. F.A.K. receivedsupport from the World Health Organization formeta-analysis of this paper. C.D.M. served as aconsultant for Otsuka Pharmaceuticals; andreceived research support from Janssen and

Unitaid. H.S.S. received research supportfrom the U.S. NIH, IMPAACT, and OtsukaPharmaceuticals. J.R.S. served on a dataand safety monitoring board for OtsukaPharmaceuticals. P.N., G.B.M., T.C., S.R.M.,B.S., J.L.B., A.C., J.P.C., L.C., T.L.D., R.D.,F.F., F.K., Z.L., A.L., M.L., J.M.M., S.M.M., L.M.,D.M., D.M.N., F.P., C.A.P., A.R., N.S.S., G.S.,J.W.W., and J.M.W reported no relevantcommercial relationships.

Acknowledgment: The authors thank LindsayMcKenna, as part of the writing committee, and

the Treatment Action Group, New York, NY, forproviding patient advocacy and communityengagement input. They also thank DennisFalzon (World Health Organization) forcontributions as an observer. They alsothank Kevin Wilson, Kimberly Lawrence,and Judy Corn (American Thoracic Society);Marc Miravitlles and Valerie Vaccaro(European Respiratory Society); PhilLoBue, Angela Starks, Tracy Ayers,Sapna Bamrah Morris (CDC); and JenniferPadberg (Infectious Diseases Society ofAmerica).

References

1. Barlam TF, Cosgrove SE, Abbo LM, MacDougall C, Schuetz AN,Septimus EJ, et al. Implementing an antibiotic stewardship program:guidelines by the Infectious Diseases Society of America and theSociety for Healthcare Epidemiology of America. Clin Infect Dis 2016;62:e51–e77.

2. U.S. Food and Drug Administration. FDA approves newdrug for treatment-resistant forms of tuberculosis that affectsthe lungs; 2019 [accessed 2019 Sep 23]. Available from:https://www.fda.gov/news-events/press-announcements/fda-approves-new-drug-treatment-resistant-forms-tuberculosis-affects-lungs.

3. Ahmad N, Ahuja SD, Akkerman OW, Alffenaar JC, Anderson LF, BaghaeiP, et al.; Collaborative Group for the Meta-Analysis of IndividualPatient Data in MDR-TB treatment–2017. Treatment correlatesof successful outcomes in pulmonary multidrug-resistanttuberculosis: an individual patient data meta-analysis. Lancet 2018;392:821–834.

4. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P,et al.; GRADE Working Group. GRADE: an emerging consensus onrating quality of evidence and strength of recommendations. BMJ2008;336:924–926.

5. Schunemann HJ, Jaeschke R, Cook DJ, Bria WF, El-Solh AA, Ernst A,et al.; ATS Documents Development and Implementation Committee.An official ATS statement: grading the quality of evidence andstrength of recommendations in ATS guidelines andrecommendations. Am J Respir Crit Care Med 2006;174:605–614.

6. Silva DR, Rendon A, Alffenaar JW, Chakaya JM, Sotgiu G, Esposito S,et al. Global TB Network: working together to eliminate tuberculosis. JBras Pneumol 2018;44:347–349.

7. World Health Organization. WHO consolidated guidelines on drug-resistant tuberculosis treatment. Geneva, Switzerland: World HealthOrganization; 2019.

8. Nunn AJ, Phillips PPJ, Meredith SK, Chiang CY, Conradie F, Dalai D,et al.; STREAM Study Collaborators. A trial of a shorter regimen forrifampin-resistant tuberculosis. N Engl J Med 2019;380:1201–1213.

9. D’Ambrosio L, Dara M, Tadolini M, Centis R, Sotgiu G, van der Werf MJ,et al.; European national programme representatives. Tuberculosiselimination: theory and practice in Europe. Eur Respir J 2014;43:1410–1420.

10. Lonnroth K, Migliori GB, Abubakar I, D’Ambrosio L, de Vries G, Diel R,et al. Towards tuberculosis elimination: an action framework for low-incidence countries. Eur Respir J 2015;45:928–952.

11. Nahid P, Dorman SE, Alipanah N, Barry PM, Brozek JL, Cattamanchi A,et al. Official American Thoracic Society/Centers for Disease Controland Prevention/Infectious Diseases Society of America ClinicalPractice Guidelines: treatment of drug-susceptible tuberculosis. ClinInfect Dis 2016;63:e147–e195.

12. World Health Organization. Global tuberculosis report 2018. Geneva,Switzerland: World Health Organization; 2018.

13. Lewinsohn DM, Leonard MK, LoBue PA, Cohn DL, Daley CL, DesmondE, et al. Official American Thoracic Society/Infectious DiseasesSociety of America/Centers for Disease Control and PreventionClinical Practice Guidelines: diagnosis of tuberculosis in adults andchildren. Clin Infect Dis 2017;64:111–115.

14. Clinical and Laboratory Standards Institute (CLSI). Susceptibility testingof mycobacteria, Nocardia spp., and other aerobic actinomycetes,3rd ed. Wayne, PA: Clinical and Laboratory Standards Institute(CLSI); 2018.

15. International Council of Nurses and Curry International TuberculosisCenter. Nursing guide for managing side effects to drug-resistant TBtreatment. 2018 [accessed 2019 Jul 15]. Available from: http://www.tbonline.info/media/uploads/documents/citc_nursingguide_english_v7_web_spreads.pdf.

16. Curry International Tuberculosis Center and California Department ofPublic Health. Drug-resistant tuberculosis: a survival guide forclinicians, 3rd ed. Oakland, CA: Curry International TuberculosisCenter and California Department of Public Health; 2016.

17. Sentinel Project on Pediatric Drug-Resistant Tuberculosis.Management of multidrug-resistant tuberculosis in children: a fieldguide, 4th ed. Boston, MA: Sentinel Project on Pediatric Drug-Resistant Tuberculosis; 2019.

18. World Health Organization. Companion handbook to the WHO guidelinesfor the programmatic management of drug-resistant tuberculosis.Geneva, Switzerland: World Health Organization; 2014.

19. Mitnick CD, White RA, Lu C, Rodriguez CA, Bayona J, Becerra MC,et al.; Collaborative Group for Analysis of Bacteriology Data in MDR-TB Treatment. Multidrug-resistant tuberculosis treatment failuredetection depends on monitoring interval and microbiologicalmethod. Eur Respir J 2016;48:1160–1170.

20. American Thoracic Society; Centers for Disease Control andPrevention; Infectious Diseases Society of America. AmericanThoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: controlling tuberculosis inthe United States. Am J Respir Crit Care Med 2005;172:1169–1227.

21. World Health Organization. Who policy on TB infection control inhealth-care facilities, congregate settings and households: Geneva,Switzerland: World Health Organization; 2009.

22. Bothamley GH, Lange C; TBNET. Infection control, genetic assessmentof drug resistance and drug susceptibility testing in the currentmanagement of multidrug/extensively-resistant tuberculosis(M/XDR-TB) in Europe: a tuberculosis network European Trialsgroup(TBNET) study. Respir Med 2017;132:68–75.

23. World Health Organization. Who treatment guidelines for drug-resistanttuberculosis 2016 update. Geneva, Switzerland: World HealthOrganization; 2016.

24. Migliori GB, Nardell E, Yedilbayev A, D’Ambrosio L, Centis R, TadoliniM, et al. Reducing tuberculosis transmission: a consensus documentfrom the world health organization regional office for Europe. EurRespir J 2019;53:1900391.

25. Melsew YA, Doan TN, Gambhir M, Cheng AC, McBryde E, Trauer JM. Riskfactors for infectiousness of patients with tuberculosis: a systematicreview and meta-analysis. Epidemiol Infect 2018;146:345–353.

26. Dharmadhikari AS, Mphahlele M, Venter K, Stoltz A, Mathebula R,Masotla T, et al. Rapid impact of effective treatment on transmissionof multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2014;18:1019–1025.

27. Jensen PA, Lambert LA, Iademarco MF, Ridzon R; CDC.Guidelines for preventing the transmission of Mycobacteriumtuberculosis in health-care settings, 2005. MMWR Recomm Rep2005;54:1–141.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e133

28. Case Management Society of America. Standards of practice for casemanagement; 2016 [accessed 2019 Sep 15]. Available from:http://www.naylornetwork.com/cmsatoday/articles/index-v3.asp?aid=400028&issueID=53653.

29. Hickam DH, Weiss JW, Guise JM, Buckley D, Motu’apuaka M, GrahamE, et al. Outpatient case management for adults with medical illnessand complex care needs. Rockville, MD: CreateSpace IndependentPublishing Platform; 2013.

30. Ross S, Goodwin N, Curry N. Case management: what is it andhow it can best be implemented. London, UK: King’s Fund;2011.

31. Klopf LC. Tuberculosis control in the New York State Department ofCorrectional Services: a case management approach. Am J InfectControl 1998;26:534–537.

32. Advisory Council for the Elimination of Tuberculosis. Essentialcomponents of a tuberculosis prevention and control program:recommendations of the Advisory Council for the Elimination ofTuberculosis. MMWR Recomm Rep 1995;44:1–16.

33. Epstein RM, Street RL Jr. The values and value of patient-centeredcare. Ann Fam Med 2011;9:100–103.

34. Migliori GB, Sotgiu G, Rosales-Klintz S, Centis R, D’Ambrosio L,Abubakar I, et al. ERS/ECDC Statement: European Unionstandards for tuberculosis care, 2017 update. Eur Respir J 2018;51:1702678.

35. TB CARE I. International standards for tuberculosis care, 3rd ed. TheHague, the Netherlands: TB CARE I.; 2018 [accessed 2014 Jul 15].Available from: http://www.who.int/tb/publications/ISTC_3rdEd.pdf.

36. Robinson JH, Callister LC, Berry JA, Dearing KA. Patient-centered careand adherence: definitions and applications to improve outcomes. JAm Acad Nurse Pract 2008;20:600–607.

37. National Tuberculosis Controllers Association. Tuberculosis nursing: acomprehensive guide to patient care, 2nd ed. Smyrna, GA: NationalTuberculosis Controllers Association; 2011.

38. Seaworth BJ. Multidrug-resistant tuberculosis. Infect Dis Clin North Am2002;16:73–105.

39. Alipanah N, Jarlsberg L, Miller C, Linh NN, Falzon D, Jaramillo E, et al.Adherence interventions and outcomes of tuberculosis treatment: asystematic review and meta-analysis of trials and observationalstudies. PLoS Med 2018;15:e1002595.

40. Muller AM, Osorio CS, Silva DR, Sbruzzi G, de Tarso P, Dalcin R.Interventions to improve adherence to tuberculosis treatment:systematic review and meta-analysis. Int J Tuberc Lung Dis 2018;22:731–740.

41. Falzon D, Timimi H, Kurosinski P, Migliori GB, Van Gemert W,Denkinger C, et al. Digital health for the End TB Strategy: developingpriority products and making them work. Eur Respir J 2016;48:29–45.

42. Sinkou H, Hurevich H, Rusovich V, Zhylevich L, Falzon D, deColombani P, et al. Video-observed treatment for tuberculosispatients in Belarus: findings from the first programmatic experience.Eur Respir J 2017;49:1602049.

43. Chuck C, Robinson E, Macaraig M, Alexander M, Burzynski J.Enhancing management of tuberculosis treatment with video directlyobserved therapy in New York City. Int J Tuberc Lung Dis 2016;20:588–593.

44. Story A, Garfein RS, Hayward A, Rusovich V, Dadu A, Soltan V, et al.Monitoring therapy compliance of tuberculosis patients by usingvideo-enabled electronic devices. Emerg Infect Dis 2016;22:538–540.

45. Zuñiga ML, Collins K, Muñoz F, Moser K, Rangel G, Cuevas-Mota J,et al. A qualitative study exploring stakeholder perceptions ofvideo directly observed therapy for monitoring tuberculosistreatment in the US-Mexico border region. J Mob Technol Med 2016;5:12–23.

46. Garfein RS, Collins K, Muñoz F, Moser K, Cerecer-Callu P, Raab F,et al. Feasibility of tuberculosis treatment monitoring by videodirectly observed therapy: a binational pilot study. Int J Tuberc LungDis 2015;19:1057–1064.

47. World Health Organization. Digital health for the End TB Strategy: anagenda for action. Geneva, Switzerland: World Health Organization;2015.

48. Velayutham B, Nair D, Ramalingam S, Perez-Velez CM, Becerra MC,Swaminathan S. Setting priorities for a research agenda to combatdrug-resistant tuberculosis in children. Public Health Action 2015;5:222–235.

49. Garfein RS, Liu L, Cuevas-Mota J, Collins K, Muñoz F, Catanzaro DG,et al. Tuberculosis treatment monitoring by video directly observedtherapy in 5 health districts, California, USA. Emerg Infect Dis 2018;24:1806–1815.

50. Schnippel K, Ndjeka N, Maartens G, Meintjes G, Master I, Ismail N,et al. Effect of bedaquiline on mortality in South African patients withdrug-resistant tuberculosis: a retrospective cohort study. LancetRespir Med 2018;6:699–706.

51. Seddon JA, Schaaf HS, Marais BJ, McKenna L, Garcia-Prats AJ,Hesseling AC, et al. Time to act on injectable-free regimens forchildren with multidrug-resistant tuberculosis. Lancet Respir Med2018;6:662–664.

52. Gunther G, Lange C, Alexandru S, Altet N, Avsar K, Bang D, et al.; forTBNET. Treatment outcomes in multidrug-resistant tuberculosis. NEngl J Med 2016;375:1103–1105.

53. Goble M, Iseman MD, Madsen LA, Waite D, Ackerson L, Horsburgh CRJr. Treatment of 171 patients with pulmonary tuberculosisresistant to isoniazid and rifampin. N Engl J Med 1993;328:527–532.

54. Falzon D, Jaramillo E, Schunemann HJ, Arentz M, Bauer M, Bayona J,et al. WHO guidelines for the programmatic management ofdrug-resistant tuberculosis: 2011 update. Eur Respir J 2011;38:516–528.

55. World Health Organization. Guidelines for the programmaticmanagement of drug-resistant tuberculosis. Geneva, Switzerland:World Health Organization; 2008.

56. Tiberi S, du Plessis N, Walzl G, Vjecha MJ, Rao M, Ntoumi F, et al.Tuberculosis: progress and advances in development of new drugs,treatment regimens, and host-directed therapies. Lancet Infect Dis2018;18:e183–e198.

57. Imperial MZ, Nahid P, Phillips PPJ, Davies GR, Fielding K, Hanna D,et al. A patient-level pooled analysis of treatment-shorteningregimens for drug-susceptible pulmonary tuberculosis. Nat Med2018;24:1708–1715.

58. Kasik JE. The nature of mycobacterial penicillinase. Am Rev Respir Dis1965;91:117–119.

59. Seung KJ, Becerra MC, Atwood SS, Alcantara F, Bonilla CA, MitnickCD. Salvage therapy for multidrug-resistant tuberculosis. ClinMicrobiol Infect 2014;20:441–446.

60. Donald PR, Sirgel FA, Venter A, Parkin DP, Van de Wal BW,Barendse A, et al. Early bactericidal activity of amoxicillin incombination with clavulanic acid in patients with sputum smear-positive pulmonary tuberculosis. Scand J Infect Dis 2001;33:466–469.

61. Chambers HF, Kocagoz T, Sipit T, Turner J, Hopewell PC. Activity ofamoxicillin/clavulanate in patients with tuberculosis. Clin Infect Dis1998;26:874–877.

62. Nadler JP, Berger J, Nord JA, Cofsky R, Saxena M. Amoxicillin-clavulanic acid for treating drug-resistant Mycobacteriumtuberculosis. Chest 1991;99:1025–1026.

63. Gonzalo X, Drobniewski F. Is there a place for b-lactams in thetreatment of multidrug-resistant/extensively drug-resistant tuberculosis? Synergy between meropenemand amoxicillin/clavulanate. J Antimicrob Chemother 2013;68:366–369.

64. Dauby N, Muylle I, Mouchet F, Sergysels R, Payen MC.Meropenem/clavulanate and linezolid treatment for extensivelydrug-resistant tuberculosis. Pediatr Infect Dis J 2011;30:812–813.

65. Sotgiu G, D’Ambrosio L, Centis R, Tiberi S, Esposito S, Dore S,et al. Carbapenems to treat multidrug and extensively drug-resistant tuberculosis: a systematic review. Int J Mol Sci 2016;17:373.

66. Tiberi S, Payen MC, Sotgiu G, D’Ambrosio L, Alarcon Guizado V,Alffenaar JW, et al. Effectiveness and safety ofmeropenem/clavulanate-containing regimens in the treatmentof MDR- and XDR-TB. Eur Respir J 2016;47:1235–1243.

AMERICAN THORACIC SOCIETY DOCUMENTS

e134 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

67. De Lorenzo S, Alffenaar JW, Sotgiu G, Centis R, D’Ambrosio L, Tiberi S,et al. Efficacy and safety of meropenem-clavulanate added tolinezolid-containing regimens in the treatment of MDR-/XDR-TB. EurRespir J 2013;41:1386–1392.

68. Swaminathan A, du Cros P, Seddon JA, Quinnell S, Bobokhojaev OI,Dusmatova Z, et al. Treating children for drug-resistant tuberculosisin Tajikistan with Group 5 medications. Int J Tuberc Lung Dis 2016;20:474–478.

69. Ahmed I, Jabeen K, Inayat R, Hasan R. Susceptibility testing ofextensively drug-resistant and pre-extensively drug-resistantMycobacterium tuberculosis against levofloxacin, linezolid, andamoxicillin-clavulanate. Antimicrob Agents Chemother 2013;57:2522–2525.

70. Gillies M, Ranakusuma A, Hoffmann T, Thorning S, McGuire T, GlasziouP, et al. Common harms from amoxicillin: a systematic review andmeta-analysis of randomized placebo-controlled trials for anyindication. CMAJ 2015;187:E21–E31.

71. Mase S, Chorba T, Lobue P, Castro K; Centers for DiseaseControl and Prevention. Provisional CDC guidelines for the useand safety monitoring of bedaquiline fumarate (Sirturo) for thetreatment of multidrug-resistant tuberculosis. MMWR Recomm Rep2013;62:1–12.

72. U.S. Food and Drug Administration. Package insert: Sirturo [accessed2019 Oct 26]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/204384s010lbl.pdf.

73. Andries K, Verhasselt P, Guillemont J, Gohlmann HW, Neefs JM,Winkler H, et al. A diarylquinoline drug active on the ATP synthase ofMycobacterium tuberculosis. Science 2005;307:223–227.

74. Almeida D, Ioerger T, Tyagi S, Li SY, Mdluli K, Andries K, et al.Mutations in pepQ confer low-level resistance to bedaquiline andclofazimine in mycobacterium tuberculosis. Antimicrob AgentsChemother 2016;60:4590–4599.

75. Nguyen TVA, Anthony RM, Bañuls AL, Nguyen TVA, Vu DH, AlffenaarJC. Bedaquiline resistance: its emergence, mechanism, andprevention. Clin Infect Dis 2018;66:1625–1630.

76. Bedaquiline (Sirturo) for multidrug-resistant tuberculosis. Med LettDrugs Ther 2013;55:66–68.

77. Pontali E, D’Ambrosio L, Centis R, Sotgiu G, Migliori GB.Multidrug-resistant tuberculosis and beyond: an updatedanalysis of the current evidence on bedaquiline. Eur Respir J2017;49:1700146.

78. Pontali E, Sotgiu G, D’Ambrosio L, Centis R, Migliori GB. Bedaquilineand multidrug-resistant tuberculosis: a systematic and criticalanalysis of the evidence. Eur Respir J 2016;47:394–402.

79. World Health Organization. Rapid communication: key changes totreatment of multidrug- and rifampicin-resistant tuberculosis(MDR/RR-TB). Geneva, Switzerland: World Health Organization;2018.

80. Borisov SE, Dheda K, Enwerem M, Romero Leyet R, D’Ambrosio L,Centis R, et al. Effectiveness and safety of bedaquiline-containingregimens in the treatment of MDR- and XDR-TB: a multicentre study.Eur Respir J 2017;49:1700387.

81. Pontali E, Sotgiu G, Tiberi S, D’Ambrosio L, Centis R, Migliori GB.Cardiac safety of bedaquiline: a systematic and critical analysis ofthe evidence. Eur Respir J 2017;50:1701462.

82. Achar J, Hewison C, Cavalheiro AP, Skrahina A, Cajazeiro J, Nargiza P,et al. Off-label use of bedaquiline in children and adolescents withmultidrug-resistant tuberculosis. Emerg Infect Dis 2017;23:10.3201/eid2310.170303.

83. Harausz EP, Garcia-Prats AJ, Seddon JA, Schaaf HS, Hesseling AC,Achar J, et al.; Sentinel Project on Pediatric Drug-ResistantTuberculosis. New and repurposed drugs for pediatric multidrug-resistant tuberculosis: practice-based recommendations. Am JRespir Crit Care Med 2017;195:1300–1310.

84. Lewis JM, Hine P, Walker J, Khoo SH, Taegtmeyer M, Squire SB, et al.First experience of effectiveness and safety of bedaquiline for 18months within an optimised regimen for XDR-TB. Eur Respir J 2016;47:1581–1584.

85. England K, Boshoff HI, Arora K, Weiner D, Dayao E, Schimel D, et al.Meropenem-clavulanic acid shows activity against Mycobacteriumtuberculosis in vivo. Antimicrob Agents Chemother 2012;56:3384–3387.

86. Hugonnet JE, Tremblay LW, Boshoff HI, Barry CE III, Blanchard JS.Meropenem-clavulanate is effective against extensively drug-resistant Mycobacterium tuberculosis. Science 2009;323:1215–1218.

87. Fox GJ, Benedetti A, Cox H, Koh WJ, Viiklepp P, Ahuja S, et al.;Collaborative Group for Meta-Analysis of Individual Patient Data inMDR-TB. Group 5 drugs for multidrug-resistant tuberculosis:individual patient data meta-analysis. Eur Respir J 2017;49:1600993.

88. Mafukidze A, Harausz E, Furin J. An update on repurposed medicationsfor the treatment of drug-resistant tuberculosis. Expert Rev ClinPharmacol 2016;9:1331–1340.

89. Tiberi S, Sotgiu G, D’Ambrosio L, Centis R, Arbex MA, AlarconArrascue E, et al. Effectiveness and safety of imipenem-clavulanateadded to an optimized background regimen (OBR) versus OBRcontrol regimens in the treatment of multidrug-resistant andextensively drug-resistant tuberculosis. Clin Infect Dis 2016;62:1188–1190.

90. Tiberi S, Sotgiu G, D’Ambrosio L, Centis R, Abdo Arbex M, AlarconArrascue E, et al. Comparison of effectiveness and safety ofimipenem/clavulanate- versus meropenem/clavulanate-containingregimens in the treatment of MDR- and XDR-TB. Eur Respir J 2016;47:1758–1766.

91. van Rijn SP, van Altena R, Akkerman OW, van Soolingen D, van der LaanT, de Lange WC, et al. Pharmacokinetics of ertapenem in patients withmultidrug-resistant tuberculosis. Eur Respir J 2016;47:1229–1234.

92. Tiberi S, D’Ambrosio L, De Lorenzo S, Viggiani P, Centis R, Sotgiu G,et al. Ertapenem in the treatment of multidrug-resistant tuberculosis:first clinical experience. Eur Respir J 2016;47:333–336.

93. Sotgiu G, Tiberi S, D’Ambrosio L, Centis R, Alffenaar JW, Caminero JA,et al.; International Carbapenem Study Group. Faster for less: thenew “shorter” regimen for multidrug-resistant tuberculosis. EurRespir J 2016;48:1503–1507.

94. Tang S, Yao L, Hao X, Liu Y, Zeng L, Liu G, et al. Clofazimine forthe treatment of multidrug-resistant tuberculosis: prospective,multicenter, randomized controlled study in China. Clin Infect Dis2015;60:1361–1367.

95. Gopal M, Padayatchi N, Metcalfe JZ, O’Donnell MR. Systematic reviewof clofazimine for the treatment of drug-resistant tuberculosis. Int JTuberc Lung Dis 2013;17:1001–1007.

96. Falzon D, Schunemann HJ, Harausz E, Gonzalez-Angulo L, LienhardtC, Jaramillo E, et al. World Health Organization treatment guidelinesfor drug-resistant tuberculosis, 2016 update. Eur Respir J 2017;49:1602308.

97. Grosset JH, Tyagi S, Almeida DV, Converse PJ, Li SY, Ammerman NC,et al. Assessment of clofazimine activity in a second-line regimenfor tuberculosis in mice. Am J Respir Crit Care Med 2013;188:608–612.

98. Lechartier B, Cole ST. Mode of action of clofazimine and combinationtherapy with benzothiazinones against mycobacterium tuberculosis.Antimicrob Agents Chemother 2015;59:4457–4463.

99. Hwang TJ, Dotsenko S, Jafarov A, Weyer K, Falzon D, Lunte K, et al.Safety and availability of clofazimine in the treatment of multidrugand extensively drug-resistant tuberculosis: analysis of publishedguidance and meta-analysis of cohort studies. BMJ Open 2014;4:e004143.

100. World Health Organization. Position statement on the continued useof the shorter MDR-TB regimen following an expedited review ofthe STREAM Stage 1 preliminary results. Geneva, Switzerland:World Health Organization; 2018.

101. Diel R, Vandeputte J, de Vries G, Stillo J, Wanlin M, Nienhaus A.Costs of tuberculosis disease in the European Union: asystematic analysis and cost calculation. Eur Respir J 2014;43:554–565.

102. Aung KJ, Van Deun A, Declercq E, Sarker MR, Das PK, Hossain MA,et al. Successful ‘9-month Bangladesh regimen’ for multidrug-resistant tuberculosis among over 500 consecutive patients. Int JTuberc Lung Dis 2014;18:1180–1187.

103. Van Deun A, Maug AK, Salim MA, Das PK, Sarker MR, Daru P, et al.Short, highly effective, and inexpensive standardized treatment ofmultidrug-resistant tuberculosis. Am J Respir Crit Care Med 2010;182:684–692.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e135

104. Piubello A, Harouna SH, Souleymane MB, Boukary I, Morou S,Daouda M, et al. High cure rate with standardised short-coursemultidrug-resistant tuberculosis treatment in Niger: no relapses. IntJ Tuberc Lung Dis 2014;18:1188–1194.

105. Kuaban C, Noeske J, Rieder HL, Aıt-Khaled N, Abena Foe JL,Trebucq A. High effectiveness of a 12-month regimen forMDR-TB patients in Cameroon. Int J Tuberc Lung Dis 2015;19:517–524.

106. Javaid A, Ahmad N, Khan AH, Shaheen Z. Applicability of the WorldHealth Organization recommended new shorter regimen in amultidrug-resistant tuberculosis high burden country. Eur Respir J2017;49:1601967.

107. van der Werf MJ, Hollo V, Kodmon C, Dara M, Catchpole M. Eligibilityfor shorter treatment of multidrug-resistant tuberculosis in theEuropean Union. Eur Respir J 2017;49:1601992.

108. Dalcolmo M, Gayoso R, Sotgiu G, D’Ambrosio L, Rocha JL, Borga L,et al. Resistance profile of drugs composing the “shorter” regimenfor multidrug-resistant tuberculosis in Brazil, 2000-2015. Eur RespirJ 2017;49:1602309.

109. Dalcolmo M, Gayoso R, Sotgiu G, D’Ambrosio L, Rocha JL, Borga L,et al. Effectiveness and safety of clofazimine in multidrug-resistanttuberculosis: a nationwide report from Brazil. Eur Respir J 2017;49:1602445.

110. Tadolini M, Lingtsang RD, Tiberi S, Enwerem M, D’Ambrosio L,Sadutshang TD, et al. Cardiac safety of extensively drug-resistanttuberculosis regimens including bedaquiline, delamanid andclofazimine. Eur Respir J 2016;48:1527–1529.

111. Wallis RS. Cardiac safety of extensively drug-resistant tuberculosisregimens including bedaquiline, delamanid and clofazimine. EurRespir J 2016;48:1526–1527.

112. Diacon AH, Dawson R, von Groote-Bidlingmaier F, Symons G,Venter A, Donald PR, et al. Bactericidal activity ofpyrazinamide and clofazimine alone and in combinations withpretomanid and bedaquiline. Am J Respir Crit Care Med 2015;191:943–953.

113. Harausz EP, Garcia-Prats AJ, Law S, Schaaf HS, Kredo T, Seddon JA,et al.; Collaborative Group for Meta-Analysis of PaediatricIndividual Patient Data in MDR-TB. Treatment and outcomes inchildren with multidrug-resistant tuberculosis: a systematic reviewand individual patient data meta-analysis. PLoS Med 2018;15:e1002591.

114. Kroger A, Pannikar V, Htoon MT, Jamesh A, Katoch K, KrishnamurthyP, et al. International open trial of uniform multi-drug therapyregimen for 6 months for all types of leprosy patients: rationale,design and preliminary results. Trop Med Int Health 2008;13:594–602.

115. Schaaf HS, Garcia-Prats AJ, McKenna L, Seddon JA. Challenges ofusing new and repurposed drugs for the treatment of multidrug-resistant tuberculosis in children. Expert Rev Clin Pharmacol 2018;11:233–244.

116. Bruning JB, Murillo AC, Chacon O, Barletta RG, Sacchettini JC.Structure of the Mycobacterium tuberculosis D-alanine:D-alanineligase, a target of the antituberculosis drug D-cycloserine.Antimicrob Agents Chemother 2011;55:291–301.

117. Cycloserine. Tuberculosis (Edinb) 2008;88:100–101.118. Hwang TJ, Wares DF, Jafarov A, Jakubowiak W, Nunn P, Keshavjee

S. Safety of cycloserine and terizidone for the treatment of drug-resistant tuberculosis: a meta-analysis. Int J Tuberc Lung Dis 2013;17:1257–1266.

119. Hemanth Kumar AK, Kumar A, Kannan T, Bhatia R, Agarwal D, KumarS, et al. Pharmacokinetics of second-line antituberculosis drugs inchildren with multidrug-resistant tuberculosis in India. AntimicrobAgents Chemother 2018;62:e02410-17.

120. Ramachandran G, Swaminathan S. Safety and tolerability profile ofsecond-line anti-tuberculosis medications. Drug Saf 2015;38:253–269.

121. Schloss J, Ismail Z. Cycloserine and isoniazid in childhoodtuberculous infections. Antibiotic Med Clin Ther (New York) 1960;7:244–248.

122. Steiner M. Newer and second-line drugs in the treatment of drug-resistant tuberculosis in children. Med Clin North Am 1967;51:1153–1167.

123. Battaglia B, Kaufman I, Lyons HA, Marsh W. Toxicity of cycloserinecombined with isoniazid in the treatment of tuberculosis in children.Am Rev Respir Dis 1961;83:751–752.

124. Ettehad D, Schaaf HS, Seddon JA, Cooke GS, Ford N. Treatmentoutcomes for children with multidrug-resistant tuberculosis: asystematic review and meta-analysis. Lancet Infect Dis 2012;12:449–456.

125. Hung WY, Yu MC, Chiang YC, Chang JH, Chiang CY, Chang CC,et al. Serum concentrations of cycloserine and outcome ofmultidrug-resistant tuberculosis in Northern Taiwan. Int J TubercLung Dis 2014;18:601–606.

126. World Health Organization. The use of delamanid in the treatment ofmultidrug-resistant tuberculosis: interim policy guidance. Geneva,Switzerland: World Health Organization; 2014.

127. von Groote-Bidlingmaier F, Patientia R, Sanchez E, Balanag V Jr,Ticona E, Segura P, et al. Efficacy and safety of delamanidin combination with an optimised background regimenfor treatment of multidrug-resistant tuberculosis: a multicentre,randomised, double-blind, placebo-controlled, parallel group phase3 trial. Lancet Respir Med 2019;7:249–259.

128. World Health Organization. WHO position statement on the use ofdelamanid for multidrug-resistant tuberculosis. Geneva,Switzerland: World Health Organization; 2018.

129. Mohr E, Hughes J, Reuter A, Trivino Duran L, Ferlazzo G, Daniels J,et al. Delamanid for rifampicin-resistant tuberculosis: aretrospective study from South Africa. Eur Respir J 2018;51:1800017.

130. Kuksa L, Barkane L, Hittel N, Gupta R. Final treatment outcomes ofmultidrug- and extensively drug-resistant tuberculosis patients inLatvia receiving delamanid-containing regimens. Eur Respir J 2017;50:1701105.

131. Centis R, Sotgiu G, Migliori GB. Delamanid: does it have arole in tuberculosis treatment? Lancet Respir Med 2019;7:193–195.

132. Lardizabal AA, Khan AN, Bamrah Morris S, Goswami ND. Notesfrom the field: acquisition of delamanid under a compassionateuse program for extensively drug-resistant tuberculosis -United States, 2017. MMWR Morb Mortal Wkly Rep 2018;67:996–997.

133. Tadolini M, Garcia-Prats AJ, D’Ambrosio L, Hewison C, Centis R,Schaaf HS, et al. Compassionate use of new drugs in children andadolescents with multidrug-resistant and extensively drug-resistanttuberculosis: early experiences and challenges. Eur Respir J 2016;48:938–943.

134. D’Ambrosio L, Centis R, Tiberi S, Tadolini M, Dalcolmo M, Rendon A,et al. Delamanid and bedaquiline to treat multidrug-resistant andextensively drug-resistant tuberculosis in children: a systematicreview. J Thorac Dis 2017;9:2093–2101.

135. World Health Organization. The use of delamanid in the treatment ofmultidrug-resistant tuberculosis in children and adolescents:interim policy guidance. Geneva, Switzerland: World HealthOrganization; 2016.

136. Dooley K, Rosenkranz S, Conradie F, Moran L, Hafner R, vonGroote-Bidlingmaier F, et al. QT effects of bedaquiline,delamanid or both in MDR-TB patients: the deliberatetrial. Presented at the Conference on Retroviruses andOpportunistic Infections (CROI). March 4–7, 2019, Seattle,WA. Abstract 84.

137. Tadolini M, Lingtsang RD, Tiberi S, Enwerem M, D’Ambrosio L,Sadutshang TD, et al. First case of extensively drug-resistanttuberculosis treated with both delamanid and bedaquiline. EurRespir J 2016;48:935–938.

138. Ferlazzo G, Mohr E, Laxmeshwar C, Hewison C, Hughes J,Jonckheere S, et al. Early safety and efficacy of the combinationof bedaquiline and delamanid for the treatment of patientswith drug-resistant tuberculosis in Armenia, India, andSouth Africa: a retrospective cohort study. Lancet Infect Dis 2018;18:536–544.

139. Kim CT, Kim TO, Shin HJ, Ko YC, Hun Choe Y, Kim HR, et al.Bedaquiline and delamanid for the treatment of multidrug-resistanttuberculosis: a multicentre cohort study in Korea. Eur Respir J2018;51:1702467.

AMERICAN THORACIC SOCIETY DOCUMENTS

e136 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

140. Pontali E, Sotgiu G, Tiberi S, Tadolini M, Visca D, D’Ambrosio L,et al. Combined treatment of drug-resistant tuberculosis withbedaquiline and delamanid: a systematic review. Eur Respir J 2018;52:1800934.

141. U.S. Food and Drug Administration. Pretomanid tablet, 200 mgmeeting of the Antimicrobial Drugs Advisory Committee (AMDAC):June 6, 2019; 2019 [accessed 2019 Jul 9]. Available from:https://www.fda.gov/advisory-committees/advisory-committee-calendar/june-6-2019-antimicrobial-drugs-advisory-committee-meeting-announcement-06062019-06062019.

142. Goude R, Amin AG, Chatterjee D, Parish T. The arabinosyltransferaseEmbC is inhibited by ethambutol in Mycobacterium tuberculosis.Antimicrob Agents Chemother 2009;53:4138–4146.

143. Donald P, Maher D, Maritz S, Qazi SA; World Health Organization.Ethambutol efficacy and toxicity: literature review andrecommendations for daily and intermittent dosage in children.Geneva, Switzerland: World Health Organization; 2006.

144. Bastos ML, Hussain H, Weyer K, Garcia-Garcia L, Leimane V, LeungCC, et al.; Collaborative Group for Meta-analysis of IndividualPatient Data in MDR-TB. Treatment outcomes of patients withmultidrug-resistant and extensively drug-resistant tuberculosisaccording to drug susceptibility testing to first- and second-linedrugs: an individual patient data meta-analysis. Clin Infect Dis 2014;59:1364–1374.

145. Cegielski JP, Dalton T, Yagui M, Wattanaamornkiet W, VolchenkovGV, Via LE, et al.; Global Preserving Effective TB Treatment Study(PETTS) Investigators. Extensive drug resistance acquired duringtreatment of multidrug-resistant tuberculosis. Clin Infect Dis 2014;59:1049–1063.

146. Graham SM, Daley HM, Banerjee A, Salaniponi FM, Harries AD.Ethambutol in tuberculosis: time to reconsider? Arch Dis Child1998;79:274–278.

147. Scardigli A, Caminero JA, Sotgiu G, Centis R, D’Ambrosio L,Migliori GB. Efficacy and tolerability of ethionamide versusprothionamide: a systematic review. Eur Respir J 2016;48:946–952.

148. Thee S, Garcia-Prats AJ, Donald PR, Hesseling AC, Schaaf HS. Areview of the use of ethionamide and prothionamide in childhoodtuberculosis. Tuberculosis (Edinb) 2016;97:126–136.

149. Ahuja SD, Ashkin D, Avendano M, Banerjee R, Bauer M, Bayona JN,et al.; Collaborative Group for Meta-Analysis of IndividualPatient Data in MDR-TB. Multidrug resistant pulmonarytuberculosis treatment regimens and patient outcomes: anindividual patient data meta-analysis of 9,153 patients. PLoS Med2012;9:e1001300.

150. Dutta BS, Hassan G, Waseem Q, Saheer S, Singh A. Ethionamide-induced hypothyroidism. Int J Tuberc Lung Dis 2012;16:141.

151. McDonnell ME, Braverman LE, Bernardo J. Hypothyroidism due toethionamide. N Engl J Med 2005;352:2757–2759.

152. Thee S, Zollner EW, Willemse M, Hesseling AC, Magdorf K, SchaafHS. Abnormal thyroid function tests in children on ethionamidetreatment. Int J Tuberc Lung Dis 2011;15:1191–1193, i.

153. Morlock GP, Metchock B, Sikes D, Crawford JT, Cooksey RC. ethA,inhA, and katG loci of ethionamide-resistant clinical Mycobacteriumtuberculosis isolates. Antimicrob Agents Chemother 2003;47:3799–3805.

154. Ling LL, Schneider T, Peoples AJ, Spoering AL, Engels I, Conlon BP,et al. A new antibiotic kills pathogens without detectable resistance.Nature 2015;517:455–459.

155. Bennett JE, Dolin R, Blaser MJ. Mandell, Douglas, and Bennett’sprinciples and practice of infectious diseases e-book. Philadelphia,PA: Elsevier Health Sciences; 2014.

156. Angeby KA, Jureen P, Giske CG, Chryssanthou E, Sturegard E,Nordvall M, et al. Wild-type MIC distributions of fourfluoroquinolones active against Mycobacterium tuberculosisin relation to current critical concentrations and availablepharmacokinetic and pharmacodynamic data. J AntimicrobChemother 2010;65:946–952.

157. Peloquin CA, Hadad DJ, Molino LP, Palaci M, Boom WH, Dietze R,et al. Population pharmacokinetics of levofloxacin, gatifloxacin, andmoxifloxacin in adults with pulmonary tuberculosis. AntimicrobAgents Chemother 2008;52:852–857.

158. Chan ED, Laurel V, Strand MJ, Chan JF, Huynh ML, Goble M, et al.Treatment and outcome analysis of 205 patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med 2004;169:1103–1109.

159. Falzon D, Gandhi N, Migliori GB, Sotgiu G, Cox HS, Holtz TH, et al.;Collaborative Group for Meta-Analysis of Individual Patient Datain MDR-TB. Resistance to fluoroquinolones and second-lineinjectable drugs: impact on multidrug-resistant TB outcomes.Eur Respir J 2013;42:156–168.

160. Cegielski JP, Kurbatova E, van der Walt M, Brand J, Ershova J,Tupasi T, et al.; Global PETTS Investigators. Multidrug-resistanttuberculosis treatment outcomes in relation to treatment and initialversus acquired second-line drug resistance. Clin Infect Dis 2016;62:418–430.

161. Seddon JA, Hesseling AC, Godfrey-Faussett P, Schaaf HS. Hightreatment success in children treated for multidrug-resistanttuberculosis: an observational cohort study. Thorax 2014;69:458–464.

162. American Society of Hospital Pharmacists (AHFS). AHFS druginformation. Bethesda, MD: Board of Directors of the AmericanSociety of Hospital Pharmacists; 2018.

163. U.S. Food and Drug Administration. FDA safety announcement: FDAwarns about increased risk of ruptures or tears in the aorta bloodvessel with fluoroquinolone antibiotics in certain patients; 2018[accessed 2019 Apr 7]. Available from: https://www.fda.gov/Drugs/DrugSafety/ucm628753.htm.

164. Bradley JS, Kauffman RE, Balis DA, Duffy CM, Gerbino PG, MaldonadoSD, et al. Assessment of musculoskeletal toxicity 5 years aftertherapy with levofloxacin. Pediatrics 2014;134:e146–e153.

165. Garcia-Prats AJ, Draper HR, Finlayson H, Winckler J, Burger A, FourieB, et al. Clinical and cardiac safety of long-term levofloxacin inchildren treated for multidrug-resistant tuberculosis. Clin Infect Dis2018;67:1777–1780.

166. Seddon JA, Hesseling AC, Finlayson H, Fielding K, Cox H, Hughes J,et al. Preventive therapy for child contacts of multidrug-resistanttuberculosis: a prospective cohort study. Clin Infect Dis 2013;57:1676–1684.

167. Stass H, Kubitza D. Effects of dairy products on the oral bioavailabilityof moxifloxacin, a novel 8-methoxyfluoroquinolone, in healthyvolunteers. Clin Pharmacokinet 2001;40:33–38.

168. Sorgel F, Kinzig M. Pharmacokinetics of gyrase inhibitors, Part 1:basic chemistry and gastrointestinal disposition. Am J Med 1993;94:44S–55S.

169. Staib AH, Beermann D, Harder S, Fuhr U, Liermann D. Absorptiondifferences of ciprofloxacin along the human gastrointestinal tractdetermined using a remote-control drug delivery device (HF-capsule). Am J Med 1989;87:66S–69S.

170. Denti P, Garcia-Prats AJ, Draper HR, Wiesner L, Winckler J, Thee S,et al. Levofloxacin population pharmacokinetics in South Africanchildren treated for multidrug-resistant tuberculosis. AntimicrobAgents Chemother 2018;62:e01521-17.

171. Mase SR, Jereb JA, Gonzalez D, Martin F, Daley CL, Fred D, et al.Pharmacokinetics and dosing of levofloxacin in children treated foractive or latent multidrug-resistant tuberculosis, Federated Statesof Micronesia and republic of the Marshall Islands. Pediatr Infect DisJ 2016;35:414–421.

172. Thee S, Garcia-Prats AJ, Draper HR, McIlleron HM, Wiesner L, CastelS, et al. Pharmacokinetics and safety of moxifloxacin in childrenwith multidrug-resistant tuberculosis. Clin Infect Dis 2015;60:549–556.

173. Srivastava S, Deshpande D, Pasipanodya J, Nuermberger E,Swaminathan S, Gumbo T. Optimal clinical doses offaropenem, linezolid, and moxifloxacin in children withdisseminated tuberculosis: Goldilocks. Clin Infect Dis 2016;63:S102–S109.

174. Peloquin CA, Phillips PPJ, Mitnick CD, Eisenach K, Patientia RF,Lecca L, et al. Increased doses lead to higher drug exposures oflevofloxacin for treatment of tuberculosis. Antimicrob AgentsChemother 2018;62:e00770-18.

175. Drusano GL, Louie A, Deziel M, Gumbo T. The crisis of resistance:identifying drug exposures to suppress amplification of resistantmutant subpopulations. Clin Infect Dis 2006;42:525–532.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e137

176. Gumbo T, Louie A, Deziel MR, Parsons LM, Salfinger M, Drusano GL.Selection of a moxifloxacin dose that suppresses drug resistancein Mycobacterium tuberculosis, by use of an in vitropharmacodynamic infection model and mathematical modeling. JInfect Dis 2004;190:1642–1651.

177. Brunton L. Knollmann B, Hilal-Dandan R. Goodman & Gilman’s thepharmacological basis of therapeutics. New York: McGraw HillMedical; 2018.

178. Kahlmeter G, Dahlager JI. Aminoglycoside toxicity - a review of clinicalstudies published between 1975 and 1982. J AntimicrobChemother 1984;13:9–22.

179. Arnold A, Cooke GS, Kon OM, Dedicoat M, Lipman M, Loyse A, et al.Adverse effects and choice between the injectable agents amikacinand capreomycin in multidrug-resistant tuberculosis. AntimicrobAgents Chemother 2017;61:e02586-16.

180. Sturdy A, Goodman A, Jose RJ, Loyse A, O’Donoghue M, Kon OM,et al. Multidrug-resistant tuberculosis (MDR-TB) treatment in theUK: a study of injectable use and toxicity in practice. J AntimicrobChemother 2011;66:1815–1820.

181. Sagwa EL, Ruswa N, Mavhunga F, Rennie T, Leufkens HG, Mantel-Teeuwisse AK. Comparing amikacin and kanamycin-inducedhearing loss in multidrug-resistant tuberculosis treatment underprogrammatic conditions in a Namibian retrospective cohort. BMCPharmacol Toxicol 2015;16:36.

182. Seddon JA, Thee S, Jacobs K, Ebrahim A, Hesseling AC, Schaaf HS.Hearing loss in children treated for multidrug-resistant tuberculosis.J Infect 2013;66:320–329.

183. Garcia-Prats AJ, Rose PC, Draper HR, Seddon JA, Norman J,McIlleron HM, et al. Effect of coadministration of lidocaine on thepain and pharmacokinetics of intramuscular amikacin in childrenwith multidrug-resistant tuberculosis: a randomized crossover trial.Pediatr Infect Dis J 2018;37:1199–1203.

184. Bardien S, Human H, Harris T, Hefke G, Veikondis R, Schaaf HS,et al. A rapid method for detection of five known mutationsassociated with aminoglycoside-induced deafness. BMC MedGenet 2009;10:2.

185. Selimoglu E. Aminoglycoside-induced ototoxicity. Curr Pharm Des2007;13:119–126.

186. Fischel-Ghodsian N. Genetic factors in aminoglycoside toxicity.Pharmacogenomics 2005;6:27–36.

187. Kranzer K, Elamin WF, Cox H, Seddon JA, Ford N, Drobniewski F. Asystematic review and meta-analysis of the efficacy and safety ofN-acetylcysteine in preventing aminoglycoside-induced ototoxicity:implications for the treatment of multidrug-resistant TB. Thorax2015;70:1070–1077.

188. Perletti G, Vral A, Patrosso MC, Marras E, Ceriani I, Willems P, et al.Prevention and modulation of aminoglycoside ototoxicity (review).Mol Med Rep 2008;1:3–13.

189. Livermore DM. Linezolid in vitro: mechanism and antibacterialspectrum. J Antimicrob Chemother 2003;51:ii9–ii16.

190. Beekmann SE, Gilbert DN, Polgreen PM; IDSA EmergingInfections Network. Toxicity of extended courses of linezolid:results of an Infectious Diseases Society of America EmergingInfections Network survey. Diagn Microbiol Infect Dis 2008;62:407–410.

191. Soriano A, Miro O, Mensa J. Mitochondrial toxicity associated withlinezolid. N Engl J Med 2005;353:2305–2306.

192. Migliori GB, Eker B, Richardson MD, Sotgiu G, Zellweger JP, SkrahinaA, et al.; TBNET Study Group. A retrospective TBNET assessmentof linezolid safety, tolerability and efficacy in multidrug-resistanttuberculosis. Eur Respir J 2009;34:387–393.

193. Sotgiu G, Pontali E, Migliori GB. Linezolid to treat MDR-/XDR-tuberculosis: available evidence and future scenarios. Eur Respir J2015;45:25–29.

194. Sotgiu G, Centis R, D’Ambrosio L, Spanevello A, Migliori GB;International Group for the study of Linezolid. Linezolid to treatextensively drug-resistant TB: retrospective data areconfirmed by experimental evidence. Eur Respir J 2013;42:288–290.

195. Lee M, Lee J, Carroll MW, Choi H, Min S, Song T, et al. Linezolid fortreatment of chronic extensively drug-resistant tuberculosis. N EnglJ Med 2012;367:1508–1518.

196. Sotgiu G, Centis R, D’Ambrosio L, Alffenaar JW, Anger HA, CamineroJA, et al. Efficacy, safety and tolerability of linezolid containingregimens in treating MDR-TB and XDR-TB: systematic review andmeta-analysis. Eur Respir J 2012;40:1430–1442.

197. Kamp J, Bolhuis MS, Tiberi S, Akkerman OW, Centis R, de Lange WC,et al. Simple strategy to assess linezolid exposure in patients withmulti-drug-resistant and extensively-drug-resistant tuberculosis. IntJ Antimicrob Agents 2017;49:688–694.

198. Bolhuis MS, Tiberi S, Sotgiu G, De Lorenzo S, Kosterink JG, van derWerf TS, et al. Is there still room for therapeutic drug monitoringof linezolid in patients with tuberculosis? Eur Respir J 2016;47:1288–1290.

199. Garcia-Prats AJ, Rose PC, Hesseling AC, Schaaf HS. Linezolid for thetreatment of drug-resistant tuberculosis in children: a review andrecommendations. Tuberculosis (Edinb) 2014;94:93–104.

200. Prieto LM, Santiago B, Del Rosal T, Carazo B, Jimenez AB, Perez-Gorricho B, et al. Spanish Paediatric TBRN. Linezolid-containingtreatment regimens for tuberculosis in children. Pediatr Infect Dis J2019;38:263–267.

201. Nagiec EE, Wu L, Swaney SM, Chosay JG, Ross DE, Brieland JK,et al. Oxazolidinones inhibit cellular proliferation via inhibition ofmitochondrial protein synthesis. Antimicrob Agents Chemother2005;49:3896–3902.

202. Bolhuis MS, Akkerman OW, Sturkenboom MGG, Ghimire S,Srivastava S, Gumbo T, et al. Linezolid-based regimens formultidrug-resistant tuberculosis (TB): a systematic review toestablish or revise the current recommended dose for TB treatment.Clin Infect Dis 2018;67:S327–S335.

203. Srivastava S, Magombedze G, Koeuth T, Sherman C, Pasipanodya JG,Raj P, et al. Linezolid dose that maximizes sterilizing effect whileminimizing toxicity and resistance emergence for tuberculosis.Antimicrob Agents Chemother 2017;61:e00751-17.

204. Song T, Lee M, Jeon HS, Park Y, Dodd LE, Dartois V, et al. Linezolidtrough concentrations correlate with mitochondrial toxicity-relatedadverse events in the treatment of chronic extensively drug-resistant tuberculosis. EBioMedicine 2015;2:1627–1633.

205. Vinh DC, Rubinstein E. Linezolid: a review of safety and tolerability. JInfect 2009;59:S59–S74.

206. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, GordinF, et al.; ATS Mycobacterial Diseases Subcommittee; AmericanThoracic Society; Infectious Disease Society of America. An officialATS/IDSA statement: diagnosis, treatment, and prevention ofnontuberculous mycobacterial diseases. Am J Respir Crit Care Med2007;175:367–416. [Published erratum appears in Am J Respir CritCare Med 175:744–745.]

207. Andini N, Nash KA. Intrinsic macrolide resistance of theMycobacterium tuberculosis complex is inducible. AntimicrobAgents Chemother 2006;50:2560–2562.

208. Nash KA. Intrinsic macrolide resistance in Mycobacterium smegmatisis conferred by a novel erm gene, erm(38). Antimicrob AgentsChemother 2003;47:3053–3060.

209. Truffot-Pernot C, Lounis N, Grosset JH, Ji B. Clarithromycin is inactiveagainst Mycobacterium tuberculosis. Antimicrob AgentsChemother 1995;39:2827–2828.

210. Bolhuis MS, van Altena R, van Soolingen D, de Lange WC, Uges DR,van der Werf TS, et al. Clarithromycin increases linezolid exposurein multidrug-resistant tuberculosis patients. Eur Respir J 2013;42:1614–1621.

211. van der Paardt AF, Wilffert B, Akkerman OW, de Lange WC, vanSoolingen D, Sinha B, et al. Evaluation of macrolides for possibleuse against multidrug-resistant Mycobacterium tuberculosis. EurRespir J 2015;46:444–455.

212. Lehmann J. Para-aminosalicylic acid in the treatment of tuberculosis.Lancet 1946;1:15.

213. Zheng J, Rubin EJ, Bifani P, Mathys V, Lim V, Au M, et al. para-Aminosalicylic acid is a prodrug targeting dihydrofolate reductasein Mycobacterium tuberculosis. J Biol Chem 2013;288:23447–23456.

214. Shane SJ, Laurie JH, Riley C, Boutilier M. Effect of combined therapy(dihydrostreptomycin and PAS) on the emergence of streptomycin-resistant strains of tubercle bacilli. N Engl J Med 1952;246:132–134.

AMERICAN THORACIC SOCIETY DOCUMENTS

e138 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

215. Friedman B, Sale L Jr, Goldman A. The effect of para-aminosalicylicacid (PAS) on the development of resistance of tubercle bacilli tostreptomycin. Am J Med Sci 1952;224:53–56.

216. Donald PR, Diacon AH. Para-aminosalicylic acid: the return of an oldfriend. Lancet Infect Dis 2015;15:1091–1099.

217. Zignol M, Cabibbe AM, Dean AS, Glaziou P, Alikhanova N, Ama C,et al. Genetic sequencing for surveillance of drug resistance intuberculosis in highly endemic countries: a multi-countrypopulation-based surveillance study. Lancet Infect Dis 2018;18:675–683.

218. Zignol M, Dean AS, Alikhanova N, Andres S, Cabibbe AM, Cirillo DM,et al. Population-based resistance of Mycobacterium tuberculosisisolates to pyrazinamide and fluoroquinolones: results from amulticountry surveillance project. Lancet Infect Dis 2016;16:1185–1192.

219. Velasquez GE, Calderon RI, Mitnick CD, Becerra MC, Huang CC,Zhang Z, et al. Pyrazinamide resistance assays and two-monthsputum culture status in patients with multidrug-resistanttuberculosis. Antimicrob Agents Chemother 2016;60:6766–6773.

220. Njire M, Tan Y, Mugweru J, Wang C, Guo J, Yew W, et al.Pyrazinamide resistance in Mycobacterium tuberculosis: reviewand update. Adv Med Sci 2016;61:63–71.

221. Cabibbe AM, Sotgiu G, Izco S, Migliori GB. Genotypic and phenotypicM. tuberculosis resistance: guiding clinicians to prescribe thecorrect regimens. Eur Respir J 2017;50:1702292.

222. Miotto P, Tessema B, Tagliani E, Chindelevitch L, Starks AM,Emerson C, et al. A standardised method for interpretingthe association between mutations and phenotypic drug resistance inMycobacterium tuberculosis. Eur Respir J 2017;50:1701354.

223. Fox W, Ellard GA, Mitchison DA. Studies on the treatment oftuberculosis undertaken by the british medical research counciltuberculosis units, 1946-1986, with relevant subsequentpublications. Int J Tuberc Lung Dis 1999;3:S231–S279.

224. Steele MA, Des Prez RM. The role of pyrazinamide in tuberculosischemotherapy. Chest 1988;94:845–850.

225. Saukkonen JJ, Cohn DL, Jasmer RM, Schenker S, Jereb JA, NolanCM, et al.; ATS (American Thoracic Society) Hepatotoxicity ofAntituberculosis Therapy Subcommittee. An official ATS statement:hepatotoxicity of antituberculosis therapy. Am J Respir Crit CareMed 2006;174:935–952.

226. Bastos ML, Lan Z, Menzies D. An updated systematic review andmeta-analysis for treatment of multidrug-resistant tuberculosis. EurRespir J 2017;49:1600803.

227. Saukkonen JJ, Powell K, Jereb JA. Monitoring for tuberculosis drughepatotoxicity: moving from opinion to evidence. Am J Respir CritCare Med 2012;185:598–599.

228. Zignol M, Floyd K. Resistance of Mycobacterium tuberculosis isolatesto pyrazinamide and fluoroquinolones: authors’ reply. Lancet InfectDis 2017;17:25.

229. Zignol M, Dean AS, Falzon D, van Gemert W, Wright A, van Deun A,et al. Twenty years of global surveillance of antituberculosis-drugresistance. N Engl J Med 2016;375:1081–1089.

230. Allix-Beguec C, Arandjelovic I, Bi L, Beckert P, Bonnet M, Bradley P,et al.; CRyPTIC Consortium and the 100,000 Genomes Project.Prediction of susceptibility to first-line tuberculosis drugs by DNAsequencing. N Engl J Med 2018;379:1403–1415.

231. Falzon D, Jaramillo E, Gilpin C, Weyer K. The role of novelapproaches and new findings in the pharmacology of tuberculosismedicines in improving treatment outcomes. Clin Infect Dis 2018;67:S365–S367.

232. de Miranda Silva C, Hajihosseini A, Myrick J, Nole J, Louie A, SchmidtS, et al. Effect of moxifloxacin plus pretomanid againstMycobacterium tuberculosis in log phase, acid phase, andnonreplicating-persister phase in an in vitro assay. AntimicrobAgents Chemother 2018;63:e01695-18.

233. de Miranda Silva C, Hajihosseini A, Myrick J, Nole J, Louie A, SchmidtS, et al. Effect of linezolid plus bedaquiline against mycobacteriumtuberculosis in log phase, acid phase, and nonreplicating-persisterphase in an in vitro assay. Antimicrob Agents Chemother 2018;62:e00856-18.

234. Ammerman NC, Swanson RV, Bautista EM, Almeida DV, Saini V,Omansen TF, et al. Impact of clofazimine dosing on treatmentshortening of the first-line regimen in a mouse model oftuberculosis. Antimicrob Agents Chemother 2018;62:e00636-18.

235. Tucker EW, Guglieri-Lopez B, Ordonez AA, Ritchie B, Klunk MH,Sharma R, et al. Noninvasive 11C-rifampin positron emissiontomography reveals drug biodistribution in tuberculous meningitis.Sci Transl Med 2018;10:eaau0965.

236. Zimmerman M, Blanc L, Chen PY, Dartois V, Prideaux B. Spatialquantification of drugs in pulmonary tuberculosis lesions by lasercapture microdissection liquid chromatography mass spectrometry(LCM-LC/MS). J Vis Exp 2018;(134):10.3791/57402.

237. Via LE, England K, Weiner DM, Schimel D, Zimmerman MD, DayaoE, et al. A sterilizing tuberculosis treatment regimen isassociated with faster clearance of bacteria in cavitary lesions inmarmosets. Antimicrob Agents Chemother 2015;59:4181–4189.

238. Chideya S, Winston CA, Peloquin CA, Bradford WZ, Hopewell PC,Wells CD, et al. Isoniazid, rifampin, ethambutol, and pyrazinamidepharmacokinetics and treatment outcomes among a predominantlyHIV-infected cohort of adults with tuberculosis from Botswana. ClinInfect Dis 2009;48:1685–1694.

239. Savic RM, Weiner M, MacKenzie WR, Engle M, Whitworth WC,Johnson JL, et al.; Tuberculosis Trials Consortium of the Centersfor Disease Control and Prevention. Defining the optimal dose ofrifapentine for pulmonary tuberculosis: exposure-responserelations from two phase II clinical trials. Clin Pharmacol Ther 2017;102:321–331.

240. Svensson EM, Svensson RJ, Te Brake LHM, Boeree MJ, Heinrich N,Konsten S, et al. The potential for treatment shortening withhigher rifampicin doses: relating drug exposure to treatmentresponse in patients with pulmonary tuberculosis. Clin Infect Dis2018;67:34–41.

241. Alsultan A, Peloquin CA. Therapeutic drug monitoring in the treatmentof tuberculosis: an update. Drugs 2014;74:839–854. [Publishederratum appears in Drugs 74:2061.]

242. Peloquin C. The role of therapeutic drug monitoring in mycobacterialinfections. Microbiol Spectr 2017;5:10.1128/microbiolspec.TNMI7-0029-2016.

243. Gumbo T, Alffenaar JC. Pharmacokinetic/pharmacodynamicbackground and methods and scientific evidence base for dosingof second-line tuberculosis drugs. Clin Infect Dis 2018;67:S267–S273.

244. Lange C, Alghamdi WA, Al-Shaer MH, Brighenti S, Diacon AH,DiNardo AR, et al. Perspectives for personalized therapy forpatients with multidrug-resistant tuberculosis. J Intern Med [onlineahead of print] 28 May 2018; DOI: 10.1111/joim.12780.

245. van Altena R, Dijkstra JA, van der Meer ME, Borjas Howard JF,Kosterink JG, van Soolingen D, et al. Reduced chance ofhearing loss associated with therapeutic drug monitoringof aminoglycosides in the treatment of multidrug-resistant tuberculosis. Antimicrob Agents Chemother 2017;61:e01400-16.

246. Deshpande D, Alffenaar JC, Koser CU, Dheda K, Chapagain ML,Simbar N, et al. D-cycloserine pharmacokinetics/pharmacodynamics,susceptibility, and dosing implications in multidrug-resistanttuberculosis: a Faustian deal. Clin Infect Dis 2018;67:S308–S316.

247. Brown AN, Drusano GL, Adams JR, Rodriquez JL, Jambunathan K,Baluya DL, et al. Preclinical evaluations to identify optimal linezolidregimens for tuberculosis therapy. MBio 2015;6:e01741-e15.

248. Pranger AD, van der Werf TS, Kosterink JGW, Alffenaar JWC. The roleof fluoroquinolones in the treatment of tuberculosis in 2019. Drugs2019;79:161–171.

249. Ahmad KF, Hamid S, du Cros P, Casas E, Khamraev A, Sikhondze W,et al. Effectiveness and safety of standardised shorter regimens formultidrug-resistant tuberculosis: individual patient data andaggregate data meta-analyses. Eur Respir J 2017;50:1700061.

250. Trebucq A, Schwoebel V, Kashongwe Z, Bakayoko A, Kuaban C,Noeske J, et al. Treatment outcome with a short multidrug-resistanttuberculosis regimen in nine African countries. Int J Tuberc LungDis 2018;22:17–25.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e139

251. Gama E, Madan J, Langley I, Girma M, Evans D, Rosen S, et al.Economic evaluation of a shortened standardised treatmentregimen of antituberculosis drugs for patients with multidrug-resistant tuberculosis (STREAM): study protocol. BMJ Open 2016;6:e014386.

252. Akkermans R. 48th Union World Conference on Lung Health. LancetRespir Med 2017;5:926–927.

253. World Health Organization. WHO position statement on the shorterMDR-TB regimen. Licence: Cc by-nc-sa 3.0 igo. 2018 [accessed2019 Sep 15]. Available from: https://www.who.int/tb/areas-of-work/drug-resistant-tb/treatment/WHOPositionStatementShorterRegimensSTREAMStage1.pdf?ua=1.

254. Lange C, Duarte R, Frechet-Jachym M, Guenther G, Guglielmetti L,Olaru ID, et al.; European MDR-TB database collaboration.Limited benefit of the new shorter multidrug-resistant tuberculosisregimen in Europe. Am J Respir Crit Care Med 2016;194:1029–1031.

255. Munoz-Torrico M, Salazar M, Millan M, Martınez OJA, NarvaezDLA, Segura Del Pilar M. Eligibility for the shorter regimen formultidrug-resistant tuberculosis in Mexico. Eur Respir J 2018;50:1702267.

256. Tsang CA, Shah N, Armstrong LR, Marks SM. Eligibility for a shortertreatment regimen for multidrug-resistant tuberculosis in the UnitedStates, 2011-2016. Clin Infect Dis 2019;ciz263.

257. Barry PM, Lowenthal P, True L, Henry L, Schack G, Wendorf K, et al.Benefit of the shorter multidrug-resistant tuberculosis treatmentregimen in California and modified eligibility criteria. Am J RespirCrit Care Med 2017;196:1488–1489.

258. Seifert M, Catanzaro D, Catanzaro A, Rodwell TC. Geneticmutations associated with isoniazid resistance inMycobacterium tuberculosis: a systematic review. PLoS One2015;10:e0119628.

259. Van Deun A, Chiang CY. Shortened multidrug-resistant tuberculosisregimens overcome low-level fluoroquinolone resistance. EurRespir J 2017;49:1700223.

260. Horne DJ, Pinto LM, Arentz M, Lin SY, Desmond E, Flores LL, et al.Diagnostic accuracy and reproducibility of WHO-endorsed phenotypicdrug susceptibility testing methods for first-line and second-lineantituberculosis drugs. J Clin Microbiol 2013;51:393–401.

261. Angra PK, Taylor TH, Iademarco MF, Metchock B, Astles JR,Ridderhof JC. Performance of tuberculosis drug susceptibilitytesting in U.S. laboratories from 1994 to 2008. J Clin Microbiol2012;50:1233–1239.

262. Resist-TB. DR-TB clinical trials progress report; 2018 [accessed 2018Jan 19]. Available from: http://www.resisttb.org/?page_id=1602.

263. Marrone MT, Venkataramanan V, Goodman M, Hill AC, Jereb JA, MaseSR. Surgical interventions for drug-resistant tuberculosis: a systematicreview and meta-analysis. Int J Tuberc Lung Dis 2013;17:6–16.

264. Naidoo R. Surgery for pulmonary tuberculosis. Curr Opin Pulm Med2008;14:254–259.

265. Pomerantz BJ, Cleveland JC Jr, Olson HK, Pomerantz M. Pulmonaryresection for multi-drug resistant tuberculosis. J Thorac CardiovascSurg 2001;121:448–453.

266. Pomerantz M, Mault JR. History of resectional surgery for tuberculosisand other mycobacterial infections. Chest Surg Clin N Am 2000;10:131–133, ix.

267. World Health Organization. The role of surgery in the treatment ofpulmonary tb and multidrug-and extensively drug-resistant tb.Geneva, Switzerland: World Health Organization; 2014.

268. Odell JA. The history of surgery for pulmonary tuberculosis. ThoracSurg Clin 2012;22:257–269.

269. Alexander GR, Biccard B. A retrospective review comparingtreatment outcomes of adjuvant lung resection for drug-resistanttuberculosis in patients with and without humanimmunodeficiency virus co-infection. Eur J Cardiothorac Surg2016;49:823–828.

270. Dara M, Sotgiu G, Zaleskis R, Migliori GB. Untreatable tuberculosis: issurgery the answer? Eur Respir J 2015;45:577–582.

271. Fox GJ, Mitnick CD, Benedetti A, Chan ED, Becerra M, Chiang CY,et al.; Collaborative Group for Meta-Analysis of Individual PatientData in MDR-TB. Surgery as an adjunctive treatment for multidrug-

resistant tuberculosis: an individual patient data metaanalysis. ClinInfect Dis 2016;62:887–895.

272. Harris RC, Khan MS, Martin LJ, Allen V, Moore DA, Fielding K, et al.;LSHTM MDR-TB surgery systematic review group. The effect ofsurgery on the outcome of treatment for multidrug-resistanttuberculosis: a systematic review and meta-analysis. BMC InfectDis 2016;16:262.

273. Borisov SE, D’Ambrosio L, Centis R, Tiberi S, Dheda K, Alffenaar JW,et al. Outcomes of patients with drug-resistant-tuberculosis treatedwith bedaquiline-containing regimens and undergoing adjunctivesurgery. J Infect 2019;78:35–39.

274. Gegia M, Winters N, Benedetti A, van Soolingen D, Menzies D. Treatmentof isoniazid-resistant tuberculosis with first-line drugs: a systematicreview and meta-analysis. Lancet Infect Dis 2017;17:223–234.

275. Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC,Friedman LN, et al.; American Thoracic Society, Centers for DiseaseControl and Prevention and the Infectious Diseases Society.American Thoracic Society/Centers for Disease Controland Prevention/Infectious Diseases Society of America:treatment of tuberculosis. Am J Respir Crit Care Med 2003;167:603–662.

276. Fregonese F, Ahuja SD, Akkerman OW, Arakaki-Sanchez D, AyakakaI, Baghaei P, et al. Comparison of different treatments for isoniazid-resistant tuberculosis: an individual patient data meta-analysis.Lancet Respir Med 2018;6:265–275.

277. Yee D, Valiquette C, Pelletier M, Parisien I, Rocher I, Menzies D.Incidence of serious side effects from first-line antituberculosisdrugs among patients treated for active tuberculosis. Am J RespirCrit Care Med 2003;167:1472–1477.

278. Weiner M, Burman W, Luo CC, Peloquin CA, Engle M, Goldberg S,et al. Effects of rifampin and multidrug resistance genepolymorphism on concentrations of moxifloxacin. AntimicrobAgents Chemother 2007;51:2861–2866.

279. Nijland HM, Ruslami R, Suroto AJ, Burger DM, Alisjahbana B, vanCrevel R, et al. Rifampicin reduces plasma concentrations ofmoxifloxacin in patients with tuberculosis. Clin Infect Dis 2007;45:1001–1007.

280. Vernon A, Burman W, Benator D, Khan A, Bozeman L. Acquiredrifamycin monoresistance in patients with HIV-related tuberculosistreated with once-weekly rifapentine and isoniazid: TuberculosisTrials Consortium. Lancet 1999;353:1843–1847.

281. Lutfey M, Della-Latta P, Kapur V, Palumbo LA, Gurner D, Stotzky G,et al. Independent origin of mono-rifampin-resistant Mycobacteriumtuberculosis in patients with AIDS. Am J Respir Crit Care Med 1996;153:837–840.

282. Isaakidis P, Casas EC, Das M, Tseretopoulou X, Ntzani EE, Ford N.Treatment outcomes for HIV and MDR-TB co-infected adults andchildren: systematic review and meta-analysis. Int J Tuberc LungDis 2015;19:969–978.

283. Gandhi NR, Andrews JR, Brust JC, Montreuil R, Weissman D, Heo M,et al. Risk factors for mortality among MDR- and XDR-TB patients ina high HIV prevalence setting. Int J Tuberc Lung Dis 2012;16:90–97.

284. Marks SM, Flood J, Seaworth B, Hirsch-Moverman Y, Armstrong L,Mase S, et al.; TB Epidemiologic Studies Consortium. Treatmentpractices, outcomes, and costs of multidrug-resistant andextensively drug-resistant tuberculosis, United States, 2005-2007.Emerg Infect Dis 2014;20:812–821.

285. World Health Organization. WHO policy on collaborative TB/HIVactivities: guidelines for national programmes and otherstakeholders. Geneva, Switzerland: World Health Organization;2012.

286. Blanc FX, Sok T, Laureillard D, Borand L, Rekacewicz C, Nerrienet E,et al.; CAMELIA (ANRS 1295–CIPRA KH001) Study Team. Earlierversus later start of antiretroviral therapy in HIV-infected adults withtuberculosis. N Engl J Med 2011;365:1471–1481.

287. Abdool Karim SS, Naidoo K, Grobler A, Padayatchi N, Baxter C, GrayAL, et al. Integration of antiretroviral therapy with tuberculosistreatment. N Engl J Med 2011;365:1492–1501.

288. Abdool Karim SS, Naidoo K, Grobler A, Padayatchi N, Baxter C, GrayA, et al. Timing of initiation of antiretroviral drugs duringtuberculosis therapy. N Engl J Med 2010;362:697–706.

AMERICAN THORACIC SOCIETY DOCUMENTS

e140 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019

289. Masur H, Brooks JT, Benson CA, Holmes KK, Pau AK, Kaplan JE;National Institutes of Health; Centers for Disease Control andPrevention; HIV Medicine Association of the Infectious DiseasesSociety of America. Prevention and treatment of opportunisticinfections in HIV-infected adults and adolescents: updatedguidelines from the Centers for Disease Control and Prevention,National Institutes of Health, and HIV Medicine Association of theInfectious Diseases Society of America. Clin Infect Dis 2014;58:1308–1311.

290. Padayatchi N, Abdool Karim SS, Naidoo K, Grobler A, Friedland G.Improved survival in multidrug-resistant tuberculosispatients receiving integrated tuberculosis and antiretroviraltreatment in the SAPiT Trial. Int J Tuberc Lung Dis 2014;18:147–154.

291. Palacios E, Franke M, Muñoz M, Hurtado R, Dallman R, Chalco K,et al. HIV-positive patients treated for multidrug-resistanttuberculosis: clinical outcomes in the HAART era. Int J Tuberc LungDis 2012;16:348–354. [Published erratum appears in Int J TubercLung Dis 16:568.]

292. Arentz M, Pavlinac P, Kimerling ME, Horne DJ, Falzon D,Schunemann HJ, et al.; ART study group. Use of anti-retroviraltherapy in tuberculosis patients on second-line anti-TB regimens: asystematic review. PLoS One 2012;7:e47370.

293. Dheda K, Shean K, Zumla A, Badri M, Streicher EM, Page-ShippL, et al. Early treatment outcomes and HIV status of patientswith extensively drug-resistant tuberculosis in South Africa:a retrospective cohort study. Lancet 2010;375:1798–1807.

294. O’Donnell MR, Padayatchi N, Kvasnovsky C, Werner L, Master I,Horsburgh CR Jr. Treatment outcomes for extensively drug-resistant tuberculosis and HIV co-infection. Emerg Infect Dis 2013;19:416–424.

295. Kvasnovsky CL, Cegielski JP, Erasmus R, Siwisa NO, Thomas K, derWalt ML. Extensively drug-resistant TB in Eastern Cape, SouthAfrica: high mortality in HIV-negative and HIV-positive patients. JAcquir Immune Defic Syndr 2011;57:146–152.

296. Satti H, McLaughlin MM, Hedt-Gauthier B, Atwood SS, Omotayo DB,Ntlamelle L, et al. Outcomes of multidrug-resistant tuberculosistreatment with early initiation of antiretroviral therapy for HIVco-infected patients in Lesotho. PLoS One 2012;7:e46943.

297. Pietersen E, Ignatius E, Streicher EM, Mastrapa B, Padanilam X,Pooran A, et al. Long-term outcomes of patients with extensivelydrug-resistant tuberculosis in South Africa: a cohort study. Lancet2014;383:1230–1239.

298. Vinnard C, Winston CA, Wileyto EP, Macgregor RR, Bisson GP.Isoniazid resistance and death in patients with tuberculousmeningitis: retrospective cohort study. BMJ 2010;341:c4451.

299. Tho DQ, Torok ME, Yen NT, Bang ND, Lan NT, Kiet VS, et al.Influence of antituberculosis drug resistance andMycobacterium tuberculosis lineage on outcome in HIV-associated tuberculous meningitis. Antimicrob AgentsChemother 2012;56:3074–3079.

300. Vinnard C, King L, Munsiff S, Crossa A, Iwata K, Pasipanodya J, et al.Long-term mortality of patients with tuberculous meningitis in NewYork City: a cohort study. Clin Infect Dis 2017;64:401–407.

301. Havlir DV, Kendall MA, Ive P, Kumwenda J, Swindells S, Qasba SS,et al.; AIDS Clinical Trials Group Study A5221. Timing ofantiretroviral therapy for HIV-1 infection and tuberculosis. N Engl JMed 2011;365:1482–1491.

302. Torok ME, Yen NT, Chau TT, Mai NT, Phu NH, Mai PP, et al. Timing ofinitiation of antiretroviral therapy in human immunodeficiency virus(HIV): associated tuberculous meningitis. Clin Infect Dis 2011;52:1374–1383.

303. Mallikaarjun S, Wells C, Petersen C, Paccaly A, Shoaf SE, Patil S,et al. Delamanid coadministered with antiretroviral drugs orantituberculosis drugs shows no clinically relevant drug-druginteractions in healthy subjects. Antimicrob Agents Chemother2016;60:5976–5985.

304. Kitahata MM, Koepsell TD, Deyo RA, Maxwell CL, Dodge WT, WagnerEH. Physicians’ experience with the acquired immunodeficiencysyndrome as a factor in patients’ survival. N Engl J Med 1996;334:701–706.

305. Thompson MA, Mugavero MJ, Amico KR, Cargill VA, Chang LW,Gross R, et al. Guidelines for improving entry into and retention incare and antiretroviral adherence for persons with HIV: evidence-based recommendations from an International Association ofPhysicians in AIDS Care panel. Ann Intern Med 2012;156:817–833,W-284, W-285, W-286, W-287, W-288, W-289, W-290, W-291,W-292, W-293, W-294.

306. Dodd PJ, Gardiner E, Coghlan R, Seddon JA. Burden of childhoodtuberculosis in 22 high-burden countries: a mathematical modellingstudy. Lancet Glob Health 2014;2:e453–e459.

307. Jenkins HE, Yuen CM, Rodriguez CA, Nathavitharana RR, McLaughlinMM, Donald P, et al. Mortality in children diagnosed withtuberculosis: a systematic review and meta-analysis. Lancet InfectDis 2017;17:285–295.

308. Dodd PJ, Sismanidis C, Seddon JA. Global burden of drug-resistanttuberculosis in children: a mathematical modelling study. LancetInfect Dis 2016;16:1193–1201.

309. Jenkins HE, Tolman AW, Yuen CM, Parr JB, Keshavjee S, Perez-VelezCM, et al. Incidence of multidrug-resistant tuberculosis disease inchildren: systematic review and global estimates. Lancet 2014;383:1572–1579.

310. Garcia-Prats AJ, Schaaf HS, Draper HR, Garcia-Cremades M,Winckler J, Wiesner L, et al. Pharmacokinetics, safety and optimaldosing of linezolid in children with multidrug-resistant tuberculosis.Int J Tuberc Lung Dis 2018;22:S593.

311. Schaaf HS, Garcia-Prats AJ, Hesseling AC, Seddon JA. Managingmultidrug-resistant tuberculosis in children: review of recentdevelopments. Curr Opin Infect Dis 2014;27:211–219.

312. Bothamley GH, Ehlers C, Salonka I, Skrahina A, Orcau A, CodecasaLR, et al. Pregnancy in patients with tuberculosis: a TBNET cross-sectional survey. BMC Pregnancy Childbirth 2016;16:304.

313. K DV, A M, S R, K AV. Multidrug-resistant tuberculosis in pregnancy. JColl Physicians Surg Pak 2007;17:637–639.

314. Oliveira HB, Mateus SH. Characterization of multidrug-resistanttuberculosis during pregnancy in Campinas, State of São Paulo,Brazil, from 1995 to 2007 [in Portuguese]. Rev Soc Bras Med Trop2011;44:627–630.

315. Palacios E, Dallman R, Muñoz M, Hurtado R, Chalco K, Guerra D, et al.Drug-resistant tuberculosis and pregnancy: treatment outcomes of 38cases in Lima, Peru. Clin Infect Dis 2009;48:1413–1419.

316. Shin S, Guerra D, Rich M, Seung KJ, Mukherjee J, Joseph K, et al.Treatment of multidrug-resistant tuberculosis during pregnancy: areport of 7 cases. Clin Infect Dis 2003;36:996–1003.

317. Tabarsi P, Baghaei P, Mirsaeidi M, Amiri M, Mansouri D, Novin A, et al.Multi-drug resistant tuberculosis in pregnancy: need for moreintensive treatment. Infection 2007;35:477–478.

318. Tabarsi P, Moradi A, Baghaei P, Marjani M, Shamaei M, Mansouri N, et al.Standardised second-line treatment of multidrug-resistant tuberculosisduring pregnancy. Int J Tuberc Lung Dis 2011;15:547–550.

319. Takashima T, Danno K, Tamura Y, Nagai T, Matsumoto T, Han Y, et al.Treatment outcome of patients with multidrug-resistant pulmonarytuberculosis during pregnancy [in Japanese]. Kekkaku 2006;81:413–418.

320. Jaspard M, Elefant-Amoura E, Melonio I, De Montgolfier I, Veziris N,Caumes E. Bedaquiline and linezolid for extensively drug-resistanttuberculosis in pregnant woman. Emerg Infect Dis 2017;23:10.3201/eid2310.161398.

321. Lessnau KD, Qarah S. Multidrug-resistant tuberculosis in pregnancy:case report and review of the literature. Chest 2003;123:953–956.

322. Nitta AT, Milligan D. Management of four pregnant women withmultidrug-resistant tuberculosis. Clin Infect Dis 1999;28:1298–1304.

323. Unlu M, Çimen P, Arı G, Şevket Dereli M. A successfully treatedsevere case of extensively drug-resistant tuberculosis duringpregnancy. Respir Case Rep 2015;4:67–71.

324. Mukherjee JS, Shin S, Furin J, Rich ML, Leandre F, Joseph JK, et al.New challenges in the clinical management of drug-resistanttuberculosis. Infect Dis Clin Pract 2002;11:329–339.

325. Khan M, Pillay T, Moodley J, Ramjee A, Padayatchi N. Pregnanciescomplicated by multidrug-resistant tuberculosis and HIV co-infection in Durban, South Africa. Int J Tuberc Lung Dis 2007;11:706–708.

AMERICAN THORACIC SOCIETY DOCUMENTS

American Thoracic Society Documents e141

326. Rohilla M, Joshi B, Jain V, Kalra J, Prasad GR. Multidrug-resistanttuberculosis during pregnancy: two case reports and review of theliterature. Case Rep Obstet Gynecol 2016;2016:1536281.

327. Van Kampenhout E, Bolhuis MS, Alffenaar JC, Oswald LM, KerstjensHA, de Lange WC, et al. Pharmacokinetics of moxifloxacin andlinezolid during and after pregnancy in a patient with multidrug-resistant tuberculosis. Eur Respir J 2017;49:1601724.

328. U.S. Food and Drug Administration. Pregnancy, lactation, andreproductive potential: labeling for human prescriptiondrug and biological products—content and format.Guidance for industry; 2014 [accessed 2019 Feb 8].Available from: https://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM425398.pdf.

329. Esmail A, Sabur NF, Okpechi I, Dheda K. Management of drug-resistanttuberculosis in special sub-populations including those with HIV co-infection, pregnancy, diabetes, organ-specific dysfunction, and in thecritically ill. J Thorac Dis 2018;10:3102–3118.

330. Gupta A, Mathad JS, Abdel-Rahman SM, Albano JD, Botgros R,Brown V, et al. Toward earlier inclusion of pregnant and postpartumwomen in tuberculosis drug trials: consensus statements from aninternational expert panel. Clin Infect Dis 2016;62:761–769.

331. American Thoracic Society. Targeted tuberculin testing and treatmentof latent tuberculosis infection. MMWR Recomm Rep 2000;49:1–51.

332. Management of persons exposed to multidrug-resistant tuberculosis.MMWR Recomm Rep 1992;41:61–71.

333. Marks SM, Mase SR, Morris SB. Systematic review, meta-analysis,and cost-effectiveness of treatment of latent tuberculosis to reduceprogression to multidrug-resistant tuberculosis. Clin Infect Dis2017;64:1670–1677.

334. Seddon J, Fred D, Amanullah F, Schaaf H, Starke J, Keshavjee S,et al. Post-exposure management of multidrug-resistanttuberculosis contacts: evidence-based recommendations. Dubai:Center for Global Health Delivery; 2015.

335. Bamrah S, Brostrom R, Dorina F, Setik L, Song R, Kawamura LM,et al. Treatment for LTBI in contacts of MDR-TB patients, FederatedStates of Micronesia, 2009-2012. Int J Tuberc Lung Dis 2014;18:912–918.

336. Denholm JT, Leslie DE, Jenkin GA, Darby J, Johnson PD, Graham SM,et al. Long-term follow-up of contacts exposed to multidrug-resistanttuberculosis in Victoria, Australia, 1995-2010. Int J Tuberc Lung Dis2012;16:1320–1325.

337. Adler-Shohet FC, Low J, Carson M, Girma H, Singh J. Management oflatent tuberculosis infection in child contacts of multidrug-resistanttuberculosis. Pediatr Infect Dis J 2014;33:664–666.

338. Williams B, Ramroop S, Shah P, Anderson L, Das S, Riddell A, et al.Management of pediatric contacts of multidrug resistant tuberculosisin the United Kingdom. Pediatr Infect Dis J 2013;32:926–927.

339. Schaaf HS, Gie RP, Kennedy M, Beyers N, Hesseling PB, Donald PR.Evaluation of young children in contact with adult multidrug-resistant pulmonary tuberculosis: a 30-month follow-up. Pediatrics2002;109:765–771.

340. Horn DL, Hewlett D Jr, Alfalla C, Peterson S, Opal SM. Limitedtolerance of ofloxacin and pyrazinamide prophylaxis againsttuberculosis. N Engl J Med 1994;330:1241.

341. Papastavros T, Dolovich LR, Holbrook A, Whitehead L, Loeb M. Adverseevents associated with pyrazinamide and levofloxacin in the treatmentof latent multidrug-resistant tuberculosis. CMAJ 2002;167:131–136.

342. Younossian AB, Rochat T, Ketterer JP, Wacker J, Janssens JP.High hepatotoxicity of pyrazinamide and ethambutol fortreatment of latent tuberculosis. Eur Respir J 2005;26:462–464.

343. Alsultan A, Peloquin CA. Clinical pharmacology of the anti-tuberculosis drugs. In: Davies PDO, Gordon SB, Davies G, editors.Clinical tuberculosis. New York, NY: CRC Press; 2014. pp.224–243.

344. Handbook of anti-tuberculosis agents: introduction. Tuberculosis(Edinb) 2008;88:85–86.

AMERICAN THORACIC SOCIETY DOCUMENTS

e142 American Journal of Respiratory and Critical Care Medicine Volume 200 Number 10 | November 15 2019