Latent tuberculosis infection latent TB -201… · Diane Havlir declared that she received research support from National Institutes of Health for research on TB. Marieke van der

Latent tuberculosis

infection Updated and consolidated

guidelines for programmatic management

Latent tuberculosis infection Updated and consolidated guidelines for

programmatic management

Latent tuberculosis infection: updated and consolidated guidelines for programmatic management

ISBN 978-92-4-155023-9

© World Health Organization 2018

Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo).

Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”.

Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization.

Suggested citation. Latent tuberculosis infection: updated and consolidated guidelines for programmatic management. Geneva: World Health Organization; 2018. Licence: CC BY-NC-SA 3.0 IGO.

Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris.

Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing.

Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user.

General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use.

WHO/CDS/TB/2018.4

iii

Contents

Acknowledgements v

Declaration and management of conflicts of interest vii

Abbreviations ix

Definitions x

Executive summary 1

1. Introduction 5

2. Identification of populations for testing and treatment of latent tuberculosis infection 9

3. Algorithms for ruling out active tuberculosis disease 14

4. Testing for latent tuberculosis infection 21

5. Treatment options for latent tuberculosis infection 23

6. Preventive treatment for contacts of patients with multidrug-resistant tuberculosis 28

7. Issues in implementation 30

8. Research priorities 32

9. References 34

Annex 1. GRADE profile tables for new recommendations 39

Annex 2. Evidence-to-Decision and GRADE tables (online at www.who.int/tb)

Annex 3. Survey on the values and preferences for the management of latent tuberculosis infection (online at www.who.int/tb)

CONTENTS

v

Acknowledgements

ACKNOWLEDGEMENTS

Overall coordination and writing of the guidelinesYohhei Hamada and Haileyesus Getahun coordinated the development and writing of the guidelines under the overall direction of Mario Raviglione.

WHO steering groupAndrei Dadu (TB and M/XDR-TB Programme, WHO Regional Office for Europe); Meg Doherty (HIV department, WHO); Dennis Falzon (Global TB Programme, WHO); Nathan Ford (HIV department, WHO); Haileyesus Getahun (Global TB Programme, WHO); Christopher Gilpin (Global TB Programme, WHO); Malgorzata Grzemska (Global TB Programme, WHO); Yohhei Hamada (Global TB Programme, WHO); Andreas Reis (Information, Evidence and Research, WHO); Wilson Were (Department of Maternal, Newborn, Child and Adolescent Health, WHO).

Co-chairs of the WHO Guidelines Development GroupAlberto Matteelli (University of Brescia, WHO Collaborative Centre for TB/HIV and TB Elimination, Italy) and Nandi Siegfried (independent clinical epidemiologist, South Africa)

GRADE methodologist for the WHO Guidelines Development GroupNandi Siegfried (independent clinical epidemiologist, South Africa).

Members of WHO Guidelines Development GroupSi Thu Aung (National TB Programme, Department of Public Health, Ministry of Health, Myanmar); Rolando Cedillos (Service of Infectology and Integrated Programme for STI/HIV/AIDS, El Salvador); Richard Chaisson (Center for TB Research, Johns Hopkins University, United States of America); Padmapriyadarsini Chandrasekaran (National Institute for Research in Tuberculosis, India); Lucy Chesire (TB Advocacy Consortium, Kenya); Betina Durovni (Federal University of Rio de Janeiro, Brazil); Diana Gibb (Medical Research Council, United Kingdom of Great Britain and Northern Ireland); Stephen Graham (Centre for International Child Health, University of Melbourne, Australia); Sally Hargreaves (Imperial College London, United Kingdom); Diane Havlir (University of California, USA); Nguyen van Hung (National Lung Hospital, Viet Nam); Mohammed Al Lawati (Consultant Physician, Oman); Alberto Matteelli (University of Brescia, WHO Collaborative Centre for TB/HIV and TB Elimination, Italy); Lindiwe Mvusi (National Department of Health, South Africa); Nadia Ismail Abu Sabrah (consultant TB physician and epidemiologist, Jordan); Marieke van der Werf (European Centre for Disease Prevention and Control, Sweden); Wim Vandevelde (Global TB Community Advisory Board, South Africa); Irina Vasilyeva (Ministry of Health, Russian Federation)

ObserversAnand Date (Centers for Disease Control and Prevention, USA); Ya Diul Mukadi (US Agency for International Development [USAID], USA).

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENTvi

Systematic reviewersYalemzewed Assefa (University of Queensland, Australia); Yibeltal Assefa (University of Queensland, Australia); Ioana Diana Olaru (Borstel Research Centre, Germany); Darshini Govindasamy (South African Medical Research Council, South Africa); Yohhei Hamada (Global TB Programme, WHO), Sandra Kik (KNCV Tuberculosis Foundation, Netherlands); Katharina Kranzer (London School of Hygiene and Tropical Medicine, United Kingdom); Johnny Lujan (Global TB Programme, WHO); Molebogeng Xheedha Rangaka (University College London, United Kingdom); Karl Schenkel (Global TB Programme, WHO); Solomon Woldeyohannes (University of Queensland, Australia).

Peer reviewers Sevim Ahmedov (USAID, USA); Taye Tolera Balcha (Armauer Hansen Research Institute, Ethiopia); Amy Bloom (USAID, USA); Anne Detjen (UNICEF); Christine Ho (Centers for Disease Control and Prevention, USA); Sundari Mase (WHO Country Office, India); Michael Kimerling (KNCV Tuberculosis Foundation, Netherlands); Giovanni Battista Migliori (WHO Collaborating Centre for TB and Lung Diseases, Fondazione S. Maugeri, Italy); Zelalem Temesgen (Mayo Clinic, USA).

Other contributors Liani Smit (University of Stellenbosch and Western Cape Department of Health, South Africa) provided input for the survey of values and preferences; Enrico Girardi and Monica Sanè Schepisi (Istituto Nazionale Malattie Infettive L. Spallanzani, Italy) shared results from an updated systematic review of the cost-effectiveness of testing and treatment for latent tuberculosis.

WHO staff and consultants Global TB Programme: Annabel Baddeley, Ernesto Jaramillo, Avinash Kanchar, Knut Lönnroth, Kefas Samson, Karin Weyer

HIV department: Satvinder Singh

Financial supportPreparation of these guidelines was supported financially by the US Centers for Disease Control and Prevention, USAID and the Ministry of Health of the Republic of Korea.

vii

Declaration and management of conflicts of interest

DECLARATION AND MANAGEMENT OF CONFLICTS OF INTEREST

All the contributors completed a WHO declaration of interests form. All the declarations were evaluated by three members of the Steering Group for any financial conflict of interest that might warrant exclusion from membership of the Guidelines Development or Peer Review Group or from discussions during preparation of the guidelines. Intellectual conflict of interest was not considered a motive for exclusion from membership of the Guidelines Development Group, as broad expertise on latent tuberculosis (TB) was considered a criterion for selection. In addition, the diversity and representation in the Group was considered large enough to balance and overcome any potential intellectual conflict of interest. The biographies of the GDG members were made public before the meeting, in line with WHO’s policy on conflicts of interest. The completed forms were reviewed at the beginning of the meeting of the Guidelines Development Group, at which time members could update their declarations.

Guidelines Development Group The following Guidelines Development Group members declared interests, which were judged to be in no conflict with the policy of WHO or the objectives of the meeting; none of the members declared commercial financial interests that were directly or indirectly related to the objectives of the meeting.

�� Padmapriyadarsini Chandrasekaran declared that her employer received research grants for the Model DOTS Project from USAID.

�� Diana Gibb declared that she was coordinating a trial centre for the TB CHAMP trial, which is funded by the Medical Research Council, the Department for International Development and the Wellcome Trust.

�� Stephen Graham declared that he had received remuneration from the Challenge TB for providing technical assistance in training and implementation of child contact screening and preventive therapy in Viet Nam between 2012 and 2016. He is currently a co-investigator in the V-QUIN and received research grants from the National Health and Medical Research Council, Australia. He is also a co-principal investigator of the DETECT Child TB (child TB diagnosis, treatment and prevention) in Uganda and received research grants from ELMA Philanthropies and the International Union Against Tuberculosis and Lung Disease.

�� Diane Havlir declared that she received research support from National Institutes of Health for research on TB.

�� Marieke van der Werf declared that she was an employee of the European Centre for Disease Prevention and Control, which has an interest in the management of LTBI. She was an opponent in defense of a PhD thesis that included studies on LTBI and received a remuneration of DKK 6741.16.

�� Wim Vandevelde declared that he was the chair of the Global TB Community Advisory Board on a voluntary basis. He stated that he represented the interests and needs of communities affected by TB.

The remaining members of the Group declared no conflicts of interests.

External Review GroupThe following External Review Group member declared interests that were judged to be in no conflict with the policy of WHO:

�� Michael Kimerling declared that his employer received grants for research on a 3-month regimen of weekly rifapentine plus isoniazid from USAID and for implementation of the regimen from UNITAID.

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENTviii

The remaining members of the Group declared no conflicts of interests.

Evidence reviewersThe researchers who did the systematic reviews of evidence (patients, intervention, comparator and outcomes, PICO) for these guidelines were:

�� PICO 1: Darshini Govindasamy, South African Medical Research Council, South Africa; Katharina Kranzer, London School of Hygiene and Tropical Medicine, United Kingdom; and Ioana Diana Olaru, Borstel Research Centre, Germany

�� PICO 2: Yohhei Hamada, Johnny Lujan, and Karl Schenkel, Global TB Programme, WHO

�� PICO 3: Yalemzewed Assefa, Yibeltal Assefa and Solomon Woldeyohannes, University of Queensland, Australia

�� PICO 4: Sandra Kik, KNCV Tuberculosis Foundation, Netherlands, and Molebogeng Xheedha Rangaka, University College London, United Kingdom

�� PICO 5: Yalemzewed Assefa, Yibeltal Assefa and Solomon Woldeyohannes, University of Queensland, Australia

�� PICO 6 and 7: Yohhei Hamada, and Karl Schenkel, Global TB Programme, WHO

These reviewers did not participate in formulating the recommendations for policy.

The following reviewers declared interests that were judged not to be in conflict with the policy of WHO or the objectives of the meeting:

�� Darshini Govindasamy declared that her institution (South African Medical Research Council) and research unit received grants from funders to conduct studies on TB.

�� Katharina Kranzer declared that she was employed as a consultant to the national reference laboratory (Borstel Research Centre). Funding for her to speak at several German universities, Gilead, Thermo Fisher Scientific and Roche, was paid to her employer. None of these activities were considered to be a significant conflict of interests in view of the topic of the review and the nature of the work she did. Furthermore, the presence of two other reviewers on the team, that selected papers and entered and analysed data, would overcome any potential conflict of interests.

The other evidence reviewers declared no conflict of interests.

All the declarations of interest are available as electronic file at the WHO Global Tuberculosis Programme.

ix

Abbreviations

ABBREVIATIONS

ART antiretroviral treatment

BCG bacille Calmette-Guérin

CI confidence interval

GDG Guideline Development Group

GRADE grading of recommendations assessment, development and evaluation

IGRA interferon-gamma release assay

IPT isoniazid preventive therapy or treatment

LTBI latent tuberculosis infection

MDR-TB multidrug-resistant tuberculosis

PICO patients, intervention, comparator and outcomes

RCT randomized controlled trial

TST tuberculin skin test

TB tuberculosis

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENTx

Definitions

Note: The definitions listed below apply to the terms as used in these guidelines. They may have different meanings in other contexts.

Adolescent: A person aged 10–19 years

Adult: A person over 19 years of age

Bacteriologically confirmed TB: TB diagnosed in a biological specimen by smear microscopy, culture or a WHO-approved molecular test such as Xpert MTB/RIF

Child: A person under 10 years

Contact: Any person who was exposed to a case of TB (see definition below)

Contact investigation: A systematic process for identifying previously undiagnosed cases of TB among the contacts of an index case. In some settings, the goal includes testing for LTBI to identify candidates for preventive treatment. Contact investigation consists of identification and prioritization and clinical evaluation.

High-TB-incidence country: A country with a WHO-estimated TB incidence rate of ≥ 100/100 000

Household contact: A person who shared the same enclosed living space as the index case for one or more nights or for frequent or extended daytime periods during the 3 months before the start of current treatment

Index case (index patient) of TB: The initially identified case of new or recurrent TB in a person of any age, in a specific household or other comparable setting in which others may have been exposed. An index case is the case on which a contact investigation is centred but is not necessarily the source case.

Infant: A child under 1 year of age

Latent tuberculosis infection (LTBI): A state of persistent immune response to stimulation by Mycobacterium tuberculosis antigens with no evidence of clinically manifest active TB. There is no gold standard test for direct identification of Mycobacterium tuberculosis infection in humans. The vast majority of infected people have no signs or symptoms of TB but are at risk for active TB disease.

Low-TB-incidence country: Country with a WHO-estimated TB incidence rate of < 100 per 100 000 population

Preventive treatment: Treatment offered to individuals who are considered to be at risk for TB disease in order to reduce that risk. Also referred to as LTBI treatment or preventive therapy

Tuberculosis (TB): The disease state due to Mycobacterium tuberculosis. In this document, commonly referred to as “active” TB or TB “disease” in order to distinguish it from LTBI

1

Executive summary

Latent tuberculosis infection (LTBI) is defined as a state of persistent immune response to stimulation by Mycobacterium tuberculosis antigens with no evidence of clinically manifest active TB. There is no gold standard test for LTBI. The WHO guidelines on LTBI consider the probability of progression to active TB disease in a specific risk group, the epidemiology and burden of TB, the availability of resources and the likelihood of a broad public health impact. Two fragmented recommendations have been made for the management of LTBI, which resulted in a number of guideline documents, posing a challenge to implementation. Therefore, several WHO Member States requested consolidated guidelines on LTBI management.

The updated, consolidated guidelines in this document respond to that request. They provide a comprehensive set of WHO recommendations for programmatic management of LTBI and the basis and rationale for national guidelines. These guidelines supersede previous WHO policy documents on the management of LTBI in people living with HIV and household contacts of people with TB and other at-risk groups. The guidelines were prepared in accordance with the requirements and recommended process of the WHO Guideline Review Committee. Seven systematic reviews were conducted to update the recommendations and make new ones. The Guideline Development Group considered the quality of the evidence, benefits and harms, values and preferences, equity, costs, acceptability and feasibility of implementation in formulating the recommendations and determining their strength.

The recommendations are presented logically according to the cascade of care for managing LTBI: identification of at-risk populations (adults and children living with HIV, HIV-negative adult and child contacts and other HIV-negative at-risk groups), ruling out active TB disease, testing for LTBI, providing treatment, monitoring adverse events, adherence and completion of treatment and monitoring and evaluation. The recommendations are categorized as: existing ones previously approved by the review committee and published, which are still valid; updated recommendations that were previously approved by the review committee but for which the evidence was reviewed, discussed with the Guidelines Development Group (GDG) and updated (including for clarity); and new recommendations. There are 10 existing, 7 updated and 7 new recommendations.

In general, the GDG reviewed the evidence from the systematic reviews and discussed each population risk group identified in detail for the prevalence of LTBI, the risk for progression to active TB and the incidence of active TB as compared with that in the general population. The GDG used the guiding principle that individual benefit outweighs risk as the mainstay of recommendations on LTBI testing and treatment. The GDG found clear evidence for the benefit of systematic testing and treatment of LTBI for people living with HIV and infants and children under 5 years of age who are household contacts of pulmonary TB patients, in all settings, regardless of the background epidemiology of TB. Similarly, they concluded that HIV-negative groups at clinical risk, such as patients initiating anti-TNF treatment, receiving dialysis, preparing for organ or haematological transplantation and those with silicosis would also benefit from testing and treatment of LTBI, regardless of the background TB epidemiology, because of their increased risk of progression to active TB disease.

The specific recommendations are given below.

A. Identification of at-risk populations for LTBI testing and treatment Adults, adolescents, children and infants living with HIV

�� Adults and adolescents living with HIV, with unknown or a positive tuberculin skin test (TST) and are unlikely to have active TB should receive preventive treatment of TB as part of a comprehensive package of HIV care. Treatment should be given to these individuals irrespective of the degree of immunosuppression

EXECUTIVE SUMMARY

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENT2

and also to those on antiretroviral treatment (ART), those who have previously been treated for TB and pregnant women. (Strong recommendation, high-quality evidence. Existing recommendation)

�� Infants aged < 12 months living with HIV who are in contact with a case of TB and are investigated for TB should receive 6 months of isoniazid preventive treatment (IPT) if the investigation shows no TB disease. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Children aged ≥ 12 months living with HIV who are considered unlikely to have TB disease on the basis of screening for symptoms and who have no contact with a case of TB should be offered 6 months of IPT as part of a comprehensive package of HIV prevention and care if they live in a setting with a high prevalence of TB. (Strong recommendation, low-quality evidence. Existing recommendation)

�� All children living with HIV who have successfully completed treatment for TB disease may receive isoniazid for an additional 6 months. (Conditional recommendation, low-quality evidence. Existing recommendation)

HIV-negative household contacts�� HIV-negative children aged < 5 years who are household contacts of people with bacteriologically confirmed

pulmonary TB and who are found not to have active TB on an appropriate clinical evaluation or according to national guidelines should be given TB preventive treatment. (Strong recommendation, high-quality evidence. Updated recommendation)

�� In countries with a low TB incidence, adults, adolescents and children who are household contacts of people with bacteriologically confirmed pulmonary TB should be systematically tested and treated for LTBI. (Strong recommendation, high–moderate-quality evidence. Existing recommendation)

�� In countries with a high TB incidence, children aged ≥ 5 years, adolescents and adults who are household contacts of people with bacteriologically confirmed pulmonary TB who are found not to have active TB by an appropriate clinical evaluation or according to national guidelines may be given TB preventive treatment. (Conditional recommendation, low-quality evidence. New recommendation)

Other HIV-negative at-risk groups�� Patients initiating anti-TNF treatment, patients receiving dialysis, patients preparing for an organ or

haematological transplant and patients with silicosis should be systematically tested and treated for LTBI. (Strong recommendation, low–very low-quality evidence. Updated recommendation)

�� In countries with a low TB incidence, systematic testing for and treatment of LTBI may be considered for prisoners, health workers, immigrants from countries with a high TB burden, homeless people and people who use illicit drugs. (Conditional recommendation, low–very low-quality evidence. Existing recommendation)

�� Systematic testing for LTBI is not recommended for people with diabetes, people with harmful alcohol use, tobacco smokers and underweight people unless they are already included in the above recommendations. (Conditional recommendation, very low-quality evidence. Existing recommendation)

B. Algorithms to rule out active TB disease�� Adults and adolescents living with HIV should be screened for TB according to a clinical algorithm. Those

who do not report any of the symptoms of current cough, fever, weight loss or night sweats are unlikely to have active TB and should be offered preventive treatment, regardless of their ART status. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Chest radiography may be offered to people living with HIV and on ART and preventive treatment given to those with no abnormal radiographic findings. (Conditional recommendation, low-quality evidence. New recommendation)

�� Adults and adolescents living with HIV who are screened for TB according to a clinical algorithm and who report any of the symptoms of current cough, fever, weight loss or night sweats may have active TB

3

and should be evaluated for TB and other diseases that cause such symptoms. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Infants and children living with HIV who have poor weight gain, fever or current cough or who have a history of contact with a case of TB should be evaluated for TB and other diseases that cause such symptoms. If the evaluation shows no TB, these children should be offered preventive treatment, regardless of their age. (Strong recommendation, low-quality evidence. Updated recommendation)

�� The absence of any symptoms of TB and the absence of abnormal chest radiographic findings may be used to rule out active TB disease among HIV-negative household contacts aged ≥ 5 years and other at-risk groups before preventive treatment. (Conditional recommendation, very low-quality evidence. New recommendation)

C. Testing for LTBI�� Either a tuberculin skin test (TST) or interferon-gamma release assay (IGRA) can be used to test for LTBI.

(Strong recommendation, very low-quality evidence. New recommendation)

�� People living with HIV who have a positive test for LTBI benefit more from preventive treatment than those who have a negative LTBI test; LTBI testing can be used, where feasible, to identify such individuals. (Strong recommendation, high-quality evidence. Existing recommendation)

�� LTBI testing by TST or IGRA is not a requirement for initiating preventive treatment in people living with HIV or child household contacts aged < 5 years. (Strong recommendation, moderate-quality evidence. Updated recommendation)

D. Treatment options for LTBI�� Isoniazid monotherapy for 6 months is recommended for treatment of LTBI in both adults and children

in countries with high and low TB incidence. (Strong recommendation, high-quality evidence. Existing recommendation)

�� Rifampicin plus isoniazid daily for 3 months should be offered as an alternative to 6 months of isoniazid monotherapy as preventive treatment for children and adolescents aged < 15 years in countries with a high TB incidence. (Strong recommendation, low-quality evidence. New recommendation)

�� Rifapentine and isoniazid weekly for 3 months may be offered as an alternative to 6 months of isoniazid monotherapy as preventive treatment for both adults and children in countries with a high TB incidence. (Conditional recommendation, moderate-quality evidence. New recommendation)

�� The following options are recommended for treatment of LTBI in countries with a low TB incidence as alternatives to 6 months of isoniazid monotherapy: 9 months of isoniazid, or a 3-month regimen of weekly rifapentine plus isoniazid, or 3–4 months of isoniazid plus rifampicin, or 3–4 months of rifampicin alone. (Strong recommendation, moderate–high-quality evidence. Existing recommendation)

�� In settings with high TB incidence and transmission, adults and adolescents living with HIV who have an unknown or a positive TST and are unlikely to have active TB disease should receive at least 36 months of IPT, regardless of whether they are receiving ART. IPT should also be given irrespective of the degree of immunosuppression, history of previous TB treatment and pregnancy. (Conditional recommendation, low-quality evidence. Existing recommendation).

E. Preventive treatment for contacts of patients with multidrug-resistant-TB�� In selected high-risk household contacts of patients with multidrug-resistant tuberculosis, preventive

treatment may be considered based on individualised risk assessment and a sound clinical justification. (Conditional recommendation, very low-quality evidence. New recommendation)

EXECUTIVE SUMMARY


Important additional considerationsAdverse events monitoring The risk for adverse events during preventive treatment must be minimized. Individuals receiving treatment for LTBI should be monitored routinely and regularly at monthly visits to health care providers. The prescribing health care provider should explain the disease process and the rationale for the treatment and emphasize the importance of completing it. People receiving treatment should be urged to contact their health care providers if they develop symptoms between visits, such as anorexia, nausea, vomiting, abdominal discomfort, persistent fatigue or weakness, dark-coloured urine, pale stools or jaundice. If a health care provider cannot be consulted at the onset of such symptoms, the patient should immediately stop treatment.

Adherence and completion of preventive treatment Adherence to the full course and completion of treatment are important determinants of clinical benefit, both to the individual and to the success of the programme. Interventions should be tailored to the specific needs of the risk groups and to the local context to ensure adherence and completion of treatment.

Programmatic management, monitoring and evaluationThe national programme should prepare a national plan for programmatic management of LTBI, including prioritizing groups identified as being at high risk on the basis of local epidemiology and the health system. They should create a conducive environment for the policy and the programme, including national and local policies and standard operating procedures to facilitate implementation of the recommendations in these guidelines. Programmatic management of LTBI should include monitoring and evaluation systems that are aligned with national systems for patient monitoring and surveillance. Appropriate recording and reporting tools should be developed, with standardized indicators.

5

1. Introduction

1.1 BackgroundLatent tuberculosis infection (LTBI) is defined as a state of persistent immune response to stimulation by Mycobacterium tuberculosis antigens with no evidence of clinically manifest active TB (1). As there is no “gold standard” test for LTBI, the global burden is not known with certainty; however, up to one third of the world’s population is estimated to be infected with M. tuberculosis (2–4), and the vast majority have no signs or symptoms of TB disease and are not infectious, although they are at risk for active TB disease and for becoming infectious. Several studies have shown that, on average, 5–10% of those infected will develop active TB disease over the course of their lives, usually within the first 5 years after initial infection (5). The risk for active TB disease after infection depends on several factors, the most important being immunological status (1).

Prevention of active TB disease by treatment of LTBI is a critical component of the WHO End TB Strategy (6). The efficacy of currently available treatments ranges from 60% to 90% (1). The potential benefit of treatment should, however, be carefully balanced against the risk for drug-related adverse events. Mass, population-wide LTBI testing and treatment are not feasible because the tests are imperfect, there are risks of serious and fatal side-effects, and the cost would be high, for an unproven public health impact. For infected individuals in population groups in which the risk for progression to active disease significantly exceeds that of the general population, however, the benefits are greater than the harm. Management of LTBI involves a comprehensive package of interventions: identifying and testing those individuals who should be tested, delivering effective, safe treatment in such a way that the majority of those starting a treatment regimen will complete it with no or minimal risk of adverse events, and monitoring and evaluation of the process.

1.2 RationaleCurrent WHO guidelines on LTBI are based on the probability that the condition will progress to active TB disease in specific risk groups, on the underlying epidemiology and burden of TB, the availability of resources and the likelihood of a broader public health impact. Therefore, management of LTBI is recommended for people living with HIV (7) and for children under 5 years who are household contacts of people with pulmonary TB (8) in all settings including those with a high TB incidence (estimated annual TB incidence rate, ≥ 100 per 100 000 population) and for adult contacts of people with TB and other clinical risk groups living in settings with a low TB incidence (estimated annual TB incidence rate < 100 per 100 000 population) (9–11). The cut-off point for defining a country as having a low or a high TB incidence was set by consensus by the previous Guideline Development Group (GDG) (9, 10). Although their recommendations led to a significant increase in preventive treatment of TB, particularly among people living with HIV, global coverage of the intervention is still very low (12). Furthermore, the fragmented recommendations resulted in a multiplicity of guideline documents, which posed challenges for smooth implementation. Therefore, several WHO Member States requested consolidated guidelines on LTBI management to ensure harmonized, smooth programmatic implementation. Increased interest has also been shown in programmatic management of LTBI as part of the End TB Strategy, including TB elimination (13).

1.3 ScopeThe present consolidated guidelines include a comprehensive set of WHO recommendations for the management of LTBI and will facilitate implementation of the End TB Strategy. They include updated recommendations from the 2011 WHO guidelines on intensified TB case-finding and isoniazid preventive

1. INTRODUCTION


therapy for people living with HIV in resource-constrained settings as well as recommendations on the use of LTBI testing. Other relevant recommendations approved by the guideline review committee are also included (Box 1). The guidelines presented here are the most recent and most comprehensive WHO guidelines for programmatic management of LTBI. They are expected to provide the basis and rationale for the development of national guidelines for LTBI management, adapted to the national and local epidemiology of TB, the availability of resources, the health infrastructure and other national and local determinants.

1.4 Target The guidelines are to be used primarily in national TB and HIV control programmes or their equivalents in ministries of health and for other policy-makers working on TB and HIV and infectious diseases. They are also appropriate for officials in other line ministries with work in the areas of health, including prison services, social services and immigration (such as ministries of justice or correctional services) and clinicians and public health practitioners working on TB, HIV, infectious diseases, prevention, child health and noncommunicable diseases such as chronic kidney disease and cancer.

1.5 Development of the guidelines In conformity with the process recommended by the WHO guideline review committee (14), three groups were established: a guideline steering group, composed of WHO staff, including from regional offices; the GDG, composed of a guideline methodologist, external content experts, national TB programme managers, academicians and representatives of patient groups and civil society, who provided input at all stages of the process; and the external review group, composed of experts with interest in LTBI, who reviewed the draft guidelines.

The steering group prepared the scoping document for the guidelines, which identified key questions in PICO (population, intervention, comparator, outcomes) format, the systematic reviews required as a basis for the recommendations and the process for preparing the guidelines. The group also considered questions on aspects of programmatic management of LTBI for which new evidence was likely to be available.

BOX 1.

Guidelines that have been updated or consolidated �� Use of tuberculosis interferon-gamma release assays (IGRAs) in low- and middle-income countries.

Geneva: World Health Organization; 2011 (http://www.who.int/tb/features_archive/policy_statement_igra_oct2011.pdf).

�� Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva: World Health Organization; 2011 (http://apps.who.int/iris/bitstream/10665/44472/1/9789241500708_ng.pdf).

�� Recommendations for investigating contacts of persons with infectious tuberculosis in low- and middle-income countries. Geneva: World Health Organization; 2012 (http://apps.who.int/iris/bitstream/10665/77741/1/9789241504492_eng.pdf).

�� Guidance for national tuberculosis programmes on the management of tuberculosis in children. Second edition. Geneva: World Health Organization; 2014 (http://apps.who.int/iris/bitstream/10665/112360/1/9789241548748_eng.pdf).

�� Recommendation on 36 months isoniazid preventive therapy to adults and adolescents living with HIV in resource-constrained and high TB and HIV-prevalence settings: 2015 update. Geneva: World Health Organization; 2015 (http://apps.who.int/iris/bitstream/10665/174052/1/9789241508872_eng.pdf).

�� Guidelines on the management of latent tuberculosis infection. Geneva: World Health Organization; 2015 (http://apps.who.int/medicinedocs/documents/s21682en/s21682en.pdf).

7

The following seven key questions were identified:

1. PICO 1: What is the prevalence of LTBI, risk of progression to active TB and cumulative prevalence of active TB among household contacts without HIV in different age groups in high TB incidence countries?

2. PICO 2: What is the accuracy of WHO symptomatic screening to exclude active TB in individuals with HIV on antiretroviral treatment (ART)?

3. PICO 3: What is the accuracy of symptomatic screening and/or chest x-ray to exclude active TB in contacts of pulmonary TB cases without HIV in high TB incidence countries?

4. PICO 4:1 Could Interferon-gamma release assays be used as an alternative to tuberculin skin tests to identify individuals most at risk of progression from LTBI to active TB in high TB incidence settings?

5. PICO 5: Should 3-month daily rifampicin plus isoniazid be offered as a preventive treatment option for children and adolescents less than 15 years of age as an alternative to 6 or 9 months isoniazid monotherapy in high TB incidence countries?

6. PICO 6: Should 3-month weekly rifapentine and isoniazid be offered as an alternative regimen to isoniazid monotherapy for treatment of LTBI in high TB incidence countries?

7. PICO 7: Should preventive treatment be recommended for contacts of patients with multidrug resistant /rifampicin resistant-TB?

A list of potential outcomes of interest for each question was circulated to all members of the GDG, who scored the importance of each outcome on a scale of 1 to 9: 1–3: not important; 4–6: important; and 7–9: critical. The average of the scores for each outcome was used to prioritize the outcome and to select the most important outcomes for each PICO question. The outcomes selected for each question and the scores for their importance are presented in Annexes 1 and 2.

Seven new or updated systematic reviews were conducted for these guidelines. The existing recommendations were discussed in the GDG and updated, as appropriate, including for clarity, as deemed necessary. Furthermore, an online survey was conducted to determine the preferences and values of affected populations for the management of LTBI (Annex 3). A series of virtual meetings, co-chaired by a technical expert and a guideline methodologist, were organized to appraise the evidence for each PICO and formulate recommendations. The “evidence-to-decision” tables developed on the “GRADEpro interface” were used to guide discussions on benefits and harm, the quality of the evidence, cost, feasibility, acceptability, equity, values and preferences (14). The recommendations and their strength were determined by the GDG on the basis of these factors. The guideline methodologist facilitated the discussions in order to reach consensus, which was defined as unanimous or majority agreement. Recommendations from existing WHO guidelines were initially assessed by the steering group and were later discussed and approved by the GDG. Valid recommendations that did not require updating were also included. The guidelines and the supporting documents were reviewed and endorsed by all GDG members. Remarks from the external review group were evaluated by the steering group for incorporation into the final version of the guidelines.

There are 10 existing, 7 updated and 7 new recommendations. The GRADE (grading of recommendations, assessment, development and evaluation) tables for the seven new recommendations are presented in Annex 1. The detailed evidence-to-decision tables for each PICO and the results of the systematic reviews are shown in Annex 2 and the survey on values and preferences in Annex 3.

1.6 Quality of evidence and strength of the recommendationsThe quality of evidence and the strength of the recommendations were assessed with the GRADE method (15). In this process, the quality of a body of evidence is defined as the degree of confidence that the estimates of effect (desirable or undesirable) are close to the actual effects of interest. The usefulness of an estimate of

1 This PICO was not included in the original scoping document submitted to the WHO Guidelines Review Committee but was added later at the suggestion of the GDG and in agreement with the Secretariat of the guidelines review committee.

1. INTRODUCTION


effect depends on the level of confidence in that estimate: the higher the quality of evidence, the more likely a strong recommendation can be made. A decision on the strength of the evidence also depends on other factors. The strength of a recommendation reflects the degree of confidence of the GDG that the desirable effects outweigh the undesirable effects. The desirable effects included beneficial health outcomes (e.g. prevention and early diagnosis of TB, reduced TB-related morbidity and mortality), a smaller burden of TB and more savings; whereas the undesirable effects include harm, a greater burden and more costs. The “burdens” included adherence to the recommendations by programmes, patients and caregivers (e.g. families), such as more frequent tests and taking additional medications.

The quality of evidence was categorized into four levels:

High We are very confident that the true effect lies close to that of the estimate of the effect.

Moderate We are moderately confident that the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low Our confidence in the effect estimate is limited: the true effect may be substantially different.

Very low We have very little confidence in the effect estimate: the true effect is likely to be substantially different.

The recommendations in these guidelines are either strong or conditional.

A strong recommendation is one for which the GDG was confident that the desirable effects of adherence would outweigh the undesirable effects. This could be either in favour of or against an intervention.

A conditional recommendation is one for which the GDG concluded that the desirable effects of adherence would probably outweigh the undesirable effects, but the GDG was not confident about the trade-off. The reasons for lack of confidence included: absence of high-quality evidence (few data to support the recommendation); imprecise estimates of benefit or harm (new evidence might change the ratio of risk to benefit); uncertainty or variation in the value of the outcomes for different individuals (applicable only to a specific group, population or setting); and small benefits or benefits that might not be worth the cost (including the cost of implementing the recommendation).

1.7 Publication, implementation, evaluation and expiryThese guidelines will be published on the WHO website in at least four languages (English, French, Spanish and Russian) and can be downloaded freely. The printed guidelines will be widely distributed at international and regional conferences and meetings of programme managers in all regions. Implementation of the recommendations will be monitored regularly in the annual data collection system of WHO Global TB Data Monitoring. WHO will update the guidelines 5 years after their publication or earlier if new evidence becomes available that necessitates a revision.

1.8 Presentation of the guidelines and recommendations The overall structure of these guidelines and the recommendations follow the logical cascade of management of LTBI: identification of at-risk populations (adults and children living with HIV, HIV-negative adult and child contacts of TB cases and other HIV-negative at-risk groups); ruling out active TB disease; testing for LTBI; providing treatment; monitoring adverse events; adherence to and completion of treatment; and monitoring and evaluation. The recommendations are categorized as existing (recommendations that were published in previous guidelines approved by the review committee and are still valid); updated (recommendations that were published in previous guidelines approved by the review committee, for which the evidence was reviewed, discussed in the GDG and updated, including for clarity); and new (recommendations that were made for the current guidelines).

9

2. Identification of populations for testing and treatment of latent tuberculosis infection

Not all individuals infected with M. tuberculosis develop active TB. It is estimated that the lifetime risk of an individual with LTBI for progression to active TB is 5–10% (5). The risk is particularly high among children under the age of 5 years and among people with compromised immunity (1). As preventive treatment entails risks and costs, preventive treatment of M. tuberculosis infection should be selectively targeted to the population groups at highest risk for progression to active TB disease, who would benefit most from treatment of LTBI.

In selecting at-risk populations for programmatic management of LTBI, consideration should be given to the epidemiology and pattern of transmission of TB in the country, so that treatment offers lasting protection. A critical component of programmatic management should therefore be a comprehensive individual clinical assessment that takes into account the balance between the risks and benefits for the individual receiving treatment.

This section describes at-risk populations for whom systematic LTBI testing and treatment is recommended.

2.1 Adults and adolescents living with HIV

Adults and adolescents living with HIV, with unknown or a positive tuberculin skin test (TST) and are unlikely to have active TB should receive preventive treatment of TB as part of a comprehensive package of HIV care. Treatment should be given to these individuals irrespective of the degree of immunosuppression and also to those on antiretroviral treatment (ART), those who have previously been treated for TB and pregnant women. (Strong recommendation, high-quality evidence. Existing recommendation)

Source of existing recommendation: Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva; World Health Organization; 2011 (http://apps.who.int/iris/bitstream/10665/44472/1/9789241500708_eng.pdf).

Summary of evidenceTB is the most frequent cause of AIDS-related deaths worldwide, despite progress in access to ART (16). TB caused about 400 000 deaths among people living with HIV in 2016, representing one third of all HIV deaths. Global data in 2016 indicated that people living with HIV were 21 times (95% confidence interval [95% CI] 16;27) more likely to develop active TB than those without HIV infection (12).

A systematic review of 12 randomized controlled trials (RCTs) of 8578 people living with HIV (17) found that preventive treatment reduced the overall risk for TB by 33% (relative effect [RR] 0.67, 95% CI 0.51;0.87) among people living with HIV. For those who were TST positive, the reduction increased to 64% (RR 0.36, 95% CI 0.22; 0.61). Although not statistically significant, the reduction was 14% among TST-negative people (RR 0.86, 95% CI 0.59; 1.26) and those of unknown TST status (RR 0.86, 95% CI 0.48; 1.52). Most of the studies in the review were, however, conducted before ART became available, and there is now increasing evidence from observational studies and RCTs of the efficacy of preventive treatment in people receiving ART. A double-blind RCT of 1329 people living with HIV and receiving ART indicated that those on ART with negative TST or IGRA benefited more from IPT than those who were TST or IGRA positive (18). An RCT of 2056 people living with HIV (19) and follow-up data (20) showed additive benefits of preventive treatment plus ART in reducing both TB incidence and overall mortality. The protective effect lasted for more than 5 years (20).

2. IDENTIFICATION OF POPULATIONS FOR TESTING AND TREATMENT OF LATENT TUBERCULOSIS INFECTION


Rationale for the recommendationThe GDG reviewed the evidence from the systematic reviews and discussed each population risk group identified in detail for the prevalence of LTBI, risk of progression to active TB and the incidence of active TB as compared with that in the general population. They concluded that the evidence shows a clear benefit of systematic testing and treatment of LTBI for people living with HIV.

Preventive treatment should be given to adults and adolescents living with HIV, regardless of their immune status and whether they are on ART, given the evidence of the additional protective effect with ART. A systematic review of studies conducted before ART became available showed the value of providing preventive treatment immediately after successful completion of TB treatment among people living with HIV in countries with high TB incidence (7, 21). Therefore, preventive treatment is recommended for people who were previously treated for TB. No evidence was found, however, for preventive treatment of people who had successfully completed treatment for multidrug-resistant (MDR) or extensively drug-resistant TB or those who are concomitantly receiving ART.

Pregnant women living with HIV are at risk for TB, which can have severe consequences for both the mother and the fetus (22). As isoniazid and rifampicin, the drugs commonly used in preventive treatment, are safe for use in pregnant women (23), pregnancy should not disqualify women living with HIV from receiving preventive treatment. Nevertheless, sound clinical judgement is required to determine the best time to provide it.

There is no evidence about repeated courses of preventive treatment, and hence no recommendation is made in the present guidelines. In settings with high TB transmission (as defined by local authorities), however, IPT for 36 months or longer is recommended conditionally (24) (see section 5). Clinical trials of repeated courses of preventive treatment are lacking and will be essential for updating these guidelines.

2.2 Infants and children living with HIV




Source of existing and updated recommendations: Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva; World Health Organization; 2011 (http://apps.who.int/iris/bitstream/10665/44472/1/9789241500708_eng.pdf).

Summary of evidence A systematic review conducted for the previous guidelines included two studies conducted in South Africa. One suggested a considerable reduction in mortality and protection against TB among HIV-infected children who received isoniazid for 6 months (25). The other RCT, however, showed no benefit of preventive treatment in HIV-infected infants with no known exposure to a TB case who were identified in the first 3–4 months of life, given rapid access to ART and carefully monitored every month for new exposure to TB or disease (26).

Few RCTs included children on ART. In one trial of 167 children on ART, the incidence of TB was lower in those given preventive treatment than in those who were not, but the difference was not statistically significant (incidence rate ratio 0.51, 95% CI 0.15;1.75) (27). A cohort study suggested an additive protective effect of preventive treatment in children receiving ART (28).

11

Rationale for the recommendations For infants aged <12 months old living with HIV, the GDG noted that preventive treatment should be given only to those infants who have a history of household contact with a person with TB and do not have TB disease according to investigations conducted in line with national guidelines because of limited data on the benefits. The GDG strongly recommended preventive treatment for children aged ≥ 12 months living with HIV, despite the low quality of the evidence, because of the clear benefits seen in adults with HIV and the high risk for active TB among people living with HIV.

The GDG noted that, although the evidence for the efficacy of preventive treatment in children on ART is limited, it is biologically plausible, given the evidence of additive effects in adults with HIV receiving ART. Thus, preventive treatment is recommended for children, regardless of whether they are on ART.

There is no evidence on the use of preventive treatment in children living with HIV after successful completion of TB treatment. Like adults, however, children living with HIV who are exposed to reinfection and recurrence of TB would benefit from preventive treatment. Therefore, based on this judgement, the GDG conditionally recommended that all children living with HIV who have been successfully treated for TB and are living in settings with high TB incidence, prevalence and transmission (as defined by national authorities) should receive a course of preventive treatment. Preventive treatment can be started immediately after the last dose of TB therapy or later, according to clinical judgement.

2.3 HIV-negative household contacts of a person with pulmonary TB

�� HIV-negative children aged < 5 years who are household contacts of people with bacteriologically confirmed pulmonary TB and who are found not to have active TB on an appropriate clinical evaluation or according to national guidelines should be given TB preventive treatment. (Strong recommendation, high-quality evidence. Updated recommendation)



Remark: Appropriate clinical evaluation should include assessment of the intensity of and risk for exposure, the risk for development of active TB and/or ascertainment of infection by testing for LTBI.

Source of existing recommendation: Guidelines on the management of latent tuberculosis Infection. Geneva; World Health Organization; 2015 (http://apps.who.int/medicinedocs/documents/s21682en/s21682en.pdf).

Summary of evidenceWe updated a systematic review conducted for the previous guidelines (9, 10), focusing on household contacts in countries with high TB incidence. The aim of the review was to determine the prevalence of LTBI, progression to active TB disease and the cumulative prevalence of active TB among household contacts, stratified by age. We added 19 studies published between 2014 and 2016. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2.

The prevalence of LTBI was higher among children and adolescents aged > 15 years and adults than in children < 5 years, who were at greatest risk for progression to active TB disease. In comparison with child household contacts < 5 years, the pooled risk ratios for progression to active TB were lower in children aged 5–15 years (0.28, 95% CI 0.12;0.65, four studies) and for those > 15 years (0.22, 95% CI 0.08;0.60, three studies). All household contacts, regardless of their age or LTBI status, were nevertheless at substantially higher risk for progression to active TB than the general population (Table 1).



Table 1. Pooled estimates of risk for active TB among household contacts stratified by age and baseline LTBI status as compared with the general population

Age (years)

LTBI-positive at baseline Regardless of baseline LTBI status

Follow-up < 12 months Follow-up < 24 months Follow-up < 12 months Follow-up < 24 months

No. of studies Risk ratio No. of

studies Risk ratio No. of studies Risk ratio No. of

studies Risk ratio

General population – 1.0 (reference) – 1.0 (reference) – 1.0 (reference) – 1.0 (reference)

0–4 2 24.3 (0.73–811.0) 3 22.9 (7.7–68.6) 3 25.9

(16.9–39.7) 5 14.8 (9.8–22.3)

5–14 2 27.1 (17.5–54.1) 3 8.2 (2.3–29.4) 3 24.1 (16.9–34.4) 5 6.3 (2.9–13.7)

≥ 15 1 30.7 (17.5–54.1) 2 13.4 (9.5–18.8) 1 24.7

(14.2–43.0) 3 11.7 (7.6–18.0)

Rationale for the recommendations The GDG noted the significantly higher risk of infants and young children < 5 years for developing active TB. Furthermore, the disease can develop rapidly in young children, and they are at greatest risk for severe and disseminated disease, associated with high morbidity and mortality. Therefore, the GDG strongly recommended preventive treatment for child household contacts aged < 5 years, regardless of the background epidemiology of TB, but only after active TB disease has been ruled out. Preventive treatment could also be considered conditionally for household contacts in other age groups, according to clinical judgement on the balance between harm and benefit for individuals and the national and local epidemiology of TB, with special consideration of on-going transmission of TB. The GDG also noted that the availability of resources and the capacity of the health infrastructure to assess the intensity and the risk of exposure and development of active TB, to weigh harm and benefit and to exclude active TB disease before initiation of treatment are important considerations in the conditionality of the recommendation.

2.4 Other HIV-negative at-risk groups

�� Patients initiating anti-TNF treatment, patients receiving dialysis, patients preparing for an organ or haematological transplant and patients with silicosis should be systematically tested and treated for LTBI. (Strong recommendation, low–very low-quality evidence. Updated recommendation)



Remark: People should be tested and treated for LTBI with strict adherence to human rights and the strongest ethical considerations. For example, positive test results or treatment for LTBI should not affect a person’s immigration status or delay the possibility of immigrating.

Source of existing recommendations: Guidelines on the management of latent tuberculosis infection. Geneva: World Health Organization; 2015 (http://apps.who.int/medicinedocs/documents/s21682en/s21682en.pdf).

Summary of evidence The GDG considered the three systematic reviews that were conducted for the previous LTBI guidelines to determine which of the 24 defined at-risk population groups should be priorities for LTBI testing and treatment for this update (9, 10). Evidence of an increased prevalence of LTBI, an increased risk of progression from LTBI

13

to active TB disease and an increased incidence of active TB was available for the following 15 risk groups: adult and child TB contacts, health care workers and students, people living with HIV, patients on dialysis, immigrants from countries with a high TB burden, patients initiating anti-TNF therapy, people who use illicit drugs, prisoners, homeless people, patients receiving an organ or haematological transplant, patients with silicosis, patients with diabetes, people with harmful use of alcohol, tobacco smokers and underweight people. An increased risk for progression to active TB was reported for 4 of the 15 groups: people living with HIV, adult and child TB contacts, patients on dialysis and underweight people. Increased risks for active TB were reported for all groups except underweight people.

Rationale for the recommendations The GDG noted that people in HIV-negative clinical risk groups, such as patients initiating anti-TNF treatment, patients on dialysis, patients preparing for organ or haematological transplant and patients with silicosis, would benefit from testing for and treatment of LTBI regardless of the background TB epidemiology, because of their increased risks for progression to active TB disease. The GDG made strong recommendations despite the low- to very low-quality evidence on the basis of its judgement that the identified at-risk population groups were at increased risk for progression to active TB disease and that the benefits of treatment outweigh the potential harm.

The GDG also concluded that the evidence for the benefits of systematic testing and treatment of LTBI might not outweigh the harm in the following population risk groups: health care workers, immigrants from countries with a high TB burden, prisoners, homeless people and people who use illicit drugs. The GDG judged, however, that the benefits might outweigh the harm to a greater extent in settings with a low TB incidence than in those with a high TB incidence because of the background TB epidemiology and the risks for transmission and reinfection. A decision to test for and treat LTBI systematically in these population groups should be made in accordance with the local TB epidemiology and context, health system structure, availability of resources and overall health priorities. Priority must be given to individuals who were recently infected with TB, as documented by conversion from negative to positive in LTBI tests (either IGRA or TST). The GDG also concluded that recent immigrants from countries with a high TB burden to one with a low burden should be prioritized. The GDG emphasized, however, that a person’s status – testing positive for LTBI or receiving LTBI treatment – should not affect the process, procedure or status of immigration.

The GDG noted the paucity of data from clinical trials on the benefits and harm of systematic LTBI testing and treatment of patients with diabetes, people with harmful use of alcohol, tobacco smokers and underweight people. They concluded that the benefits of systematic, routine testing and treatment of these risk groups do not outweigh the risks, unless they are part of the groups covered in the recommendations, regardless of background TB epidemiology.

The GDG agreed that prioritization of groups on the basis of their risk and the local and national context (e.g. epidemiology, resource availability) would be acceptable to individuals and to key stakeholders, including clinicians and programme managers. It noted that the high risk for ongoing TB transmission of certain risk groups, such as health care workers (including students), prisoners (and prison staff), immigrants from high-incidence countries, homeless people and people who use illicit drugs, requires attention, so that the benefit of treatment is not compromised by reinfection.

Considerations for implementation A decision of national TB programmes and other stakeholders on priority risk groups for programmatic management of LTBI should primarily take into account the evidence in providing a lasting prevention from progression to active TB (e.g. absence of continuous transmission and reinfection) with benefits outweighing risks to the individuals belong to the group and the efficient use of resources. The GDG noted that prioritization of at-risk groups for LTBI testing and treatment could yield savings for the health care system. The GDG noted the value of ART for preventing TB in people living with HIV, underlining the importance of ensuring provision of ART for all people living with HIV as per the current WHO policy (29).



3. Algorithms for ruling out active tuberculosis disease

This section recommends algorithms for ruling out active TB disease before providing preventive treatment.

3.1 Adults, adolescents, children and infants living with HIV3.1.1 Adults and adolescents with HIV

�� Adults and adolescents living with HIV should be screened for TB according to a clinical algorithm. Those who do not report any of the symptoms of current cough, fever, weight loss or night sweats are unlikely to have active TB and should be offered preventive treatment, regardless of their ART status. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Chest radiography may be offered to people living with HIV and on ART and preventive treatment be given to those with no abnormal radiographic findings. (Conditional recommendation, low-quality evidence. New recommendation)

�� Adults and adolescents living with HIV who are screened for TB according to a clinical algorithm and who report any of the symptoms of current cough, fever, weight loss or night sweats may have active TB and should be evaluated for TB and other diseases that cause such symptoms. (Strong recommendation, moderate-quality evidence. Updated recommendation)

Remark: Chest radiography should not be a requirement for initiating preventive treatment.

Summary of evidence Active TB disease must be excluded before initiating preventive treatment. In 2011, WHO conducted a systematic review and a meta-analysis of individual patient data and recommended a symptom-screening rule of a combination of current cough, weight loss, night sweats and fever in order to exclude active TB. The review showed that the rule had a sensitivity of 79%, a specificity of 50% and a negative predictive value of 97.7% at a TB prevalence of 5%. Most of the people living with HIV who were included in studies in the systematic review were not receiving ART (30).

During updating of the guidelines, we conducted a systematic review to assess the performance of the four-symptom screening rule in people living with HIV who were and were not receiving ART; 17 studies with such information were retained. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2.

The pooled sensitivity of the four-symptom screening rule for people living with HIV on ART was 51.0% (95% CI 28.4;73.2), and the specificity was 70.7% (95% CI 47.7;86.4); the pooled sensitivity of the rule for people living with HIV but not receiving ART was 89.3% (95% CI 82.6;93.6), and the specificity was 27.2% (95% CI 17.3;40.0). Two studies provided data on addition of abnormal chest radiographic findings to the screening rule for people living with HIV on ART. The pooled sensitivity was higher (84.6%, 95% CI 69.7; 92.9), but the specificity was lower (29.8%, 95% CI 26.3;33.6) than those of the symptom screen alone.

In all studies, the median prevalence of TB among people living with HIV on ART was 1.5% (interquartile range, 0.6–3.5%). At a 1% prevalence of TB, the negative predictive value of the symptom screening rule was 99.3%; addition of abnormal chest radiographic findings increased the negative predictive value by 0.2%.

No studies of the addition of chest radiography to the symptom rule for pregnant women were found in the review. The algorithm for TB screening in adults and adolescents living with HIV is shown in Fig. 1.

15

Rationale for the recommendations Overall, the GDG agreed that the four-symptom screening rule is very useful for ruling out active TB before providing preventive treatment to people living with HIV, regardless of whether they receive ART. It noted the potential benefits of adding an abnormal chest radiographic finding to the four-symptom screening rule, while recognizing a marginal improvement in performance. Moreover, increased use of chest radiography would add more false-positive results to the screening rule, which would require more investigations for TB and other illnesses. Therefore, the GDG reiterated that chest radiography should be added as an additional investigation only if it does not pose a barrier to the provision of preventive treatment for people living with HIV. Although no study was found of the additive role of chest radiography in testing pregnant women, the GDG noted that pregnant women living with HIV could also benefit, as long as good clinical practices are observed to prevent any significant risk to the fetus (31).

3. ALGORITHMS FOR RULING OUT ACTIVE TUBERCULOSIS

Fig. 1 Algorithm for screening adults and adolescents living with HIV for TB

a Every adult and adolescent should be evaluated for eligibility to receive ART. Infection control measures should be prioritized to reduce M. tuberculosis transmission in all settings in which care is provided.

b Chest radiography can be done if available, particularly for people living with HIV on ART, but is not required to classify patients into TB and non-TB groups. In settings with a high HIV prevalence and a high TB prevalence among people living with HIV (e.g. > 10%), strong consideration should be given to adding other, sensitive investigations.

c Contraindications include: active hepatitis (acute or chronic), regular and heavy alcohol consumption and symptoms of peripheral neuropathy. History of TB and current pregnancy should not be contraindications for starting preventive treatment. Although LTBI testing is not a requirement for initiating preventive treatment, it may be done as a part of eligibility screening where feasible.

d Xpert MTB/RIF should be used as the initial diagnostic test for TB. Detailed algorithms for people living with HIV suspected of having TB are available in the WHO consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection (http://apps.who.int/iris/bitstream/10665/208825/1/9789241549684_eng.pdf).

e Resume regular screening for TB after completion of treatment for active disease.

Adults and adolescents living with HIVa

Screen people with any of the following symptoms of TB:b Current cough

FeverWeight loss

Night sweats

Assess for contraindications to preventive treatmentc Investigate for TB and other diseasesd

YesNo Not TBOther diagnosis TB

Give preventive treatment

Defer preventive treatment (also screen for TB at

follow-up)

Give appropriate treatment and

consider preventive treatment

Follow up and


Treat for TBe

Screen for TB regularly, at each encounter with a health worker or visit to a health facility

No Yes


3.1.2 Children living with HIV

Infants and children living with HIV who have poor weight gain, fever or current cough or who have a history of contact with a case of TB should be evaluated for TB and other diseases that cause such symptoms. If the evaluation shows no TB, these children should be offered preventive treatment, regardless of their age. (Strong recommendation, low-quality evidence. Updated recommendation)

Poor weight gain is defined as reported weight loss, very low weight-for-age (< –3 z-score), underweight (weight-for-age < –2 z-score), confirmed weight loss (> 5%) since the last visit or growth curve flattening,

Children and infants < 1 year of age should be given preventive treatment only if they have a history of household contact with a TB case and active TB has been excluded in investigations.

Summary of evidence Infants and children living with HIV should be screened for TB routinely, as part of standard clinical care, regardless of whether they are receiving TB prophylaxis or ART. A systematic review conducted for the previous guidelines identified limited evidence on the best approach to screening infants and children for TB (7). On the basis of these few studies and expert opinion, the previous guidelines group recommended a screening rule consisting of poor weight gain, fever, current cough and a history of contact with a TB case. We conducted a new systematic review to assess the performance of this screening rule. The only publication found was the conference abstract of a study of 176 hospitalized children with HIV aged ≤ 12 years in Kenya (32). The study had a sensitivity of 100% (95% CI, 76.8;100.0) and a specificity of 4.3% (95% CI, 1.8;8.7).

Rationale for the recommendation The GDG noted the paucity of data on the usefulness of the screening rule for children living with HIV. The single study showed that the symptom screening rule currently recommended for children with HIV performs well, but no study has been reported on the harm or challenges of the rule, such as resource requirements for implementation. Symptom-based screening is generally accepted by clients and is feasible in resource-constraint settings.

Therefore, the GDG decided to make the same strong recommendation. Children living with HIV who screen positive for TB may have TB and should be evaluated for TB and other diseases. If the evaluation shows no TB, children with HIV should be offered preventive treatment, regardless of their age, while only infants < 12 months of age who have a history of household contact with a person with TB and have been investigated for TB according to national guidelines should receive preventive treatment. The GDG also noted that clinicians should broaden the differential diagnosis to include other diseases that may cause current cough, fever and poor weight gain in children with HIV. The algorithm for TB screening in children aged ≥ 1 year living with HIV is shown in Fig. 2.

3.2 HIV-negative infants, children and adults who are household contacts of a person with pulmonary TB

3.2.1 HIV-negative infants and children < 5 years of age Summary of evidenceWe updated a previous systematic review of screening algorithms for HIV-negative people and people with unknown HIV status (33). The review revealed only one study of young children (mean age, 19.2 months) in which various symptoms were evaluated, such as failure to thrive and prolonged cough. The report did not discuss the combination of symptoms for excluding TB (34).

Fig. 3 is a simple algorithm for ruling out active TB in child household contacts aged < 5 years before preventive treatment (8). The algorithm can be used by health workers at peripheral level. Research on symptom-based screening of children’s TB contacts indicates that this contact management strategy is safe and more feasible in resource-limited settings than contact screening based on diagnoses (35, 36). Furthermore, a recent modelling

17

study suggested that providing preventive treatment without LTBI testing is cost-effective for child contacts under 5 years of age (37).

3.2.2 HIV-negative household contacts aged ≥ 5 years and other at-risk groups

The absence of any symptoms of TB and the absence of abnormal chest radiographic findings may be used to rule out active TB disease among HIV-negative household contacts aged ≥ 5 years and other at-risk groups before preventive treatment. (Conditional recommendation, very low-quality evidence. New recommendation)

Summary of evidenceWe updated the previous systematic review used for the 2015 guidelines (9, 10) to determine the sensitivity and specificity of screening based on symptoms and/or chest radiography for ruling out active TB in HIV-

Fig. 2 Algorithm for screening children aged ≥ 1 year living with HIV for TB

Children aged ≥ 12 months living with HIVa

Screen for any of the following symptoms of TB: Poor weight gainb

FeverCurrent cough

History of contact with a person with TB

Assess for contraindications to preventive treatmentc Investigate for TB and other diseasesd

YesNo Not TBOther diagnosis TB

Give preventive treatment

Defer preventive treatment

Give appropriate treatment and


Follow up and


Treat for TBe

Screen for TB regularly, at each encounter with a health worker or visit to a health facility

No Yes

a All infants < 1 year of age should be given preventive treatment if they have a history of household contact with a TB case.b Poor weight gain is defined as reported weight loss, very low weight-for-age (< –3 z-score), underweight (weight-for-age < –2 z-score),

confirmed weight loss (> 5%) since the last visit or growth curve flattening. c Contraindications include active hepatitis (acute or chronic) and symptoms of peripheral neuropathy. A history of TB should not be

a contraindication for starting preventive treatment. Although LTBI testing is not a requirement for initiating preventive treatment, it may be done as a part of eligibility screening where feasible.

d Xpert MTB/RIF should be used as the initial diagnostic test for TB. Detailed algorithms for people living with HIV who have suspected TB are available in the WHO consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection (http://apps.who.int/iris/bitstream/10665/208825/1/9789241549684_eng.pdf).

e Resume regular screening for TB after completion of treatment for active disease.



negative people and people of unknown HIV status. The review covered 33 studies, including 17 newly identified. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2. To illustrate how the various screening and diagnostic algorithms are expected to rule out active TB, a simple model was constructed to compare the following six screening criteria: (i) any TB symptom, (ii) any cough, (iii) cough for > 2–3 weeks, (iv) chest radiographic abnormality suggestive of TB, (v) any chest radiographic abnormality and (vi) a combination of any chest radiographic abnormality or any TB symptom. The model suggested that the combination of any chest radiographic abnormality and the presence of any symptoms suggestive of TB (i.e. any cough of any duration, haemoptysis, fever, night sweats, weight loss, chest pain, shortness of breath and fatigue) would offer the highest sensitivity (100%) and negative predictive value (100%) for ruling out TB.

Fig. 4 presents the algorithm for targeted diagnosis and treatment of LTBI and exclusion of active TB in house-hold contacts aged ≥5 years and other at-risk populations.

Rationale for the recommendation on HIV-negative household contactsThe GDG noted the lack of any new data and agreed to continue use of the existing symptom-based algorithms (Fig. 3) for HIV-negative infants and children who are household contacts of TB cases. The GDG reiterated that active TB should be ruled out primarily by investigations, according to national guidelines. It noted that screening of child contacts could include LTBI testing and chest radiography, although the absence of those investigations should not pose a barrier for either diagnosis of active TB disease or provision of preventive treatment. In the absence of these tests, clinical assessment alone is sufficient to decide on eligibility for preventive treatment particularly for children < 5 years of age who are household contacts of a bacteriologically confirmed pulmonary TB.

a Children must be free of any TB and non-TB related symptoms to be considered well. b The commonest TB-related symptoms are persistence of: cough, fever, not eating well/anorexia, weight loss/failure to thrive, fatigue,

reduced playfulness and decreased activity. If TB disease is suspected, refer to Chapter 3 in the Guidance for national tuberculosis programmes on the management of tuberculosis in children. Second edition

(www.who.int/tb/publications/childtb_guidelines/en/).c Isoniazid 10 mg/kg (7−15 mg/kg) daily for 6 months. 3-month weekly regimen of rifapentine plus isoniazid or 3-month daily

rifampicin plus isoniazid may be used as an alternative (see section 5).d If TB disease is diagnosed, anti-TB treatment should be started and the child registered with the national TB programme. If TB disease

is excluded, the child should be considered for eligibility for preventive treatment.

Fig. 3 Algorithm for screening HIV-negative infants and children < 5 years of age who are household contacts of people with TB

Children < 5 years of age who are household contacts of people with bacteriologically confirmed pulmonary TB

Wella Symptomaticb

Preventive treatment

If become symptomatic

Evaluate for TB disease

19

The GDG also noted that symptom screening with or without the addition of chest radiography should be acceptable for individuals and programme managers. Chest radiography could increase the confidence of health care providers that active TB has been ruled out, thus reducing any concern about development of drug resistance; however, addition of chest radiography could incur costs for clients and inconvenience, as more clients would have to be investigated for TB and other diseases. The GDG noted the difficulty of assessing the preferences of children, which were not systematically documented.

3.3 Considerations for implementation Addition of abnormal chest radiographic findings to the symptom screening rule would complicate logistics, increasing the cost, workload, infrastructure and availability of qualified staff. The GDG noted that chest radiography should not be a requirement or a barrier for initiating TB preventive treatment in people living with HIV because of the need for additional resources, in view of the marginal gain in negative predictive value.

People living with HIV who have any of the four symptoms or abnormal chest radiographic findings may have active TB and should be investigated for TB and other diseases. Xpert MTB/RIF should be used as the initial diagnostic test. Other diseases that cause any of the four symptoms should be investigated in accordance with

Ask for any symptoms of tuberculosis in individuals from the risk groupa

Yes No

TST or IGRA

NegativebPositive

Chest radiography

No abnormality

Treat for LTBI

Any abnormalityTB and other disease investigationsc

Fig. 4 Algorithm for targeted diagnosis and treatment of LTBI and exclusion of active TB in HIV-negative household contacts aged ≥ 5 years and other at-risk populations

a Any symptom of TB: cough, haemoptysis, fever, night sweats, weight loss, chest pain, shortness of breath, fatigue. An HIV test could be offered according to national or local guidelines or clinical judgement. Similarly, chest radiographs could be performed at this stage.

b Clients for whom LTBI treatment is not indicated should be given information about TB, including the importance of seeking care if symptoms of TB develop.

c National TB guidelines should be followed in investigating TB. In addition, people in whom TB is excluded after investigations (including those with fibrotic radiological lesions) can be considered for LTBI treatment.



national guidelines and sound clinical practice. People living with HIV who present any of the four symptoms but in whom active TB is excluded by investigations may be considered for preventive treatment.

The four-symptom screening method is recommended for all people living with HIV at every visit to a health facility or contact with a health worker. As combining chest radiography with symptom screening at every visit could represent a significant burden on the health system as well as on clients, it should be used only to exclude active TB before giving preventive treatment, with due respect for good clinical practice. The role of chest radiography in regular TB screening and its optimal frequency is uncertain. Local authorities should define its application and frequency on the basis of their local epidemiology, health infrastructure and resource availability.

It is critical to ensure proper follow-up and investigation for TB and other diseases in household contacts with abnormal chest radiographic findings or TB symptoms. The investigations should be performed in accordance with national guidelines and sound clinical practice. Contacts in whom active TB is excluded after investigations may be considered for preventive treatment.

Chest radiography and trained health care workers (e.g. radiologists) must be available to implement the screening rule. Where chest radiography is not available, absence of any TB symptoms alone may be considered as a criteria before preventive treatment. This would offer the highest sensitivity among symptom-based screening rules, and its negative predictive value is high in most settings.

21

4. Testing for latent tuberculosis infection

4. TESTING FOR LATENT TUBERCULOSIS INFECTION

�� Either a tuberculin skin test (TST) or interferon-gamma release assay (IGRA) can be used to test for LTBI. (Strong recommendation, very low-quality evidence. New recommendation)

Remark: The availability and affordability of the tests will determine which will be chosen by clinicians and programme managers. Neither TST nor IGRA can be used to diagnose active TB disease nor for diagnostic workup of adults suspected of having active TB.



Source of existing recommendation: Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva; World Health Organization; 2011 (http://apps.who.int/iris/bitstream/10665/44472/1/9789241500708_eng.pdf).

Summary of evidence There is no gold standard method for diagnosing LTBI. TST and IGRA require a competent immune response in order to identify people infected with TB and are imperfect tests for measuring progression to active disease.

A previous systematic review was updated to compare the predictive performance of IGRA and TST for identifying incident active TB in countries with a high TB incidence (38). Only studies in which TST was compared with IGRA in the same population (“head-to-head” studies) were included. Relative risk ratios for TB for people who tested positive and those who tested negative with TST and IGRA were estimated. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2.

Five prospective cohort studies were identified, with a total of 7769 participants; four were newly identified. Three of the studies were conducted in South Africa and two in India (18, 39–42). The studies included people living with HIV, pregnant women, adolescents, health care workers and household contacts. The pooled risk ratio estimate for TST was 1.49 (95% CI, 0.79; 2.80), and that for IGRA was 2.03 (95% CI, 1.18; 3.50). Although the estimate for IGRA was slightly higher than that for TST, the 95% CIs for the estimates for TST and IGRA overlapped and were imprecise. Furthermore, there was limited evidence for the predictive utility of the tests in specific at-risk populations.

Rationale for the recommendations The evidence reviewed and the recommendations apply to the use of the two commercially available IGRAs (QuantiFERON®-TB Gold In-Tube and T-SPOT®.TB) only. The GDG concluded that the comparison of TST and IGRA in the same population does not provide strong evidence that one test should be preferred over the other for predicting progression to active TB disease. The GDG noted that TST may require significantly fewer resources than IGRA and may be more familiar to practitioners in resource-constrained settings; however, recurrent global shortages and stock-outs of TST reduce its use in scaling up programmatic management of LTBI.


The GDG also noted that equity and access could affect the choice and type of test used. The preferences of clients and programmes are, however, affected by several factors, such as the requirement for sophisticated laboratory infrastructure (e.g. for IGRA) and possible additional costs for clients (e.g. for travel) and programmes (e.g. for building and testing). The GDG strongly recommended the two tests as equivalent options, with relatively similar advantages and disadvantages.

The GDG stressed that the global shortage of TST should be addressed urgently and called for more investment into research on novel tests for LTBI with better predictive value.

The GDG cautioned that imperfect performance of these tests can lead to false-negative results, particularly for young children and immunocompromised individuals such as people living with HIV. The GDG noted the importance of the tests for identifying recent conversion from a negative to a positive result, particularly among contacts of people with pulmonary TB, which is good practice for initiating TB preventive treatment. Nevertheless, recent studies among health care workers tested serially for LTBI in the USA showed that conversions from negative to positive and reversions from positive to negative are more commonly identified with IGRA than with TST (43). Thus, sound clinical judgement must be used in interpreting the results of these tests when used serially.

The GDG recommended that LTBI testing should not be a requirement for initiating TB preventive treatment for people living with HIV and child household contacts aged < 5 years, particularly in countries with a high TB incidence, given that clear benefits outweigh the risks. HIV-negative infant and child household contacts aged < 5 years and people living with HIV who have a negative LTBI test should be assessed case by case for their individual risk of exposure to TB and the added advantage of receiving preventive treatment.

Considerations for implementation The GDG noted that the availability and affordability of the tests could determine which LTBI test is used. Other considerations include the structure of the health system, feasibility of implementation and infrastructure requirements.

The incremental cost-effectiveness of IGRAs and TSTs appears to be influenced mainly by their accuracy. Bacille Calmette-Guérin (BCG) vaccination plays a decisive role in reducing the specificity of TST, leading the choice towards IGRA-only strategies. The GDG noted, however, that the impact of BCG vaccination on the specificity of TST depends on the strain of vaccine used, the age at which the vaccine is given and the number of doses administered. When BCG is given at birth, as is the case in most parts of the world, it has a variable, limited impact on TST specificity (44). Therefore, the GDG agreed that a history of BCG vaccination has a limited effect on interpretation of TST results later in life; hence, BCG vaccination should not be a determining factor in selecting a test.

IGRAs are more costly and more technically complex to perform than the TST. Operational difficulties should be considered in deciding which test to use. For example, IGRA requires a phlebotomy, which can be difficult, particularly in very young children, laboratory infrastructure, technical expertise and expensive equipment; however, only a single visit is required to obtain a result (although patients may have to make a second visit to learn the result). TST is less costly and can be performed in the field, but it requires a cold chain, two health care visits and training in intradermal injection, reading and interpretation.

23

5. Treatment options for latent tuberculosis infection

5. TREATMENT OPTIONS FOR LATENT TUBERCULOSIS INFECTION

�� Isoniazid monotherapy for 6 months is recommended for treatment of LTBI in both adults and children in countries with high and low TB incidence. (Strong recommendation, high-quality evidence. Existing recommendation)




Remark: Rifampicin- and rifapentine-containing regimens should be prescribed with caution to people living with HIV who are on ART because of potential drug–drug interactions.

�� In settings with high TB incidence and transmission, adults and adolescents living with HIV who have an unknown or a positive TST and are unlikely to have active TB disease should receive at least 36 months of IPT, regardless of whether they are receiving ART. IPT should also be given irrespective of the degree of immunosuppression, history of previous TB treatment and pregnancy. (Conditional recommendation, low-quality evidence. Existing recommendation)

Remark: People living with HIV in settings with high TB incidence and transmission, regardless of their TST status, obtain more benefit from IPT for ≥ 36 months than for 6 months. Those with a positive TST have a greater protective benefit and those receiving ART have a significant additional benefit from longer-term IPT. Use of TST is encouraged whenever feasible, but it is not a prerequisite for IPT. People with a negative TST should not receive 36 months of IPT. Settings with high TB incidence and transmission should be defined by national authorities, taking into consideration the local epidemiology and transmission of both TB and HIV.

Source of existing recommendations : Guidelines on the management of latent tuberculosis infection. Geneva: World Health Organization; 2015 (http://apps.who.int/medicinedocs/documents/s21682en/s21682en.pdf).

Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. Geneva; World Health Organization; 2011 (http://apps.who.int/iris/bitstream/10665/44472/1/9789241500708_eng.pdf).

Guidance for national tuberculosis programmes on the management of tuberculosis in children. 2nd Edition. Geneva: World Health Organization; 2014 (http://apps.who.int/iris/bitstream/10665/112360/1/9789241548748_eng.pdf).

Recommendation on 36 months isoniazid preventive therapy to adults and adolescents living with HIV in resource-constrained and high TB- and HIV-prevalence settings: 2015 update. Geneva: World Health Organization; 2015 (http://apps.who.int/iris/bitstream/10665/174052/1/9789241508872_eng.pdf).


Summary of evidence

5.1 Daily isoniazid monotherapy6 months’ isoniazid monotherapySix months’ daily monotherapy with isoniazid is the standard treatment for both adults and children living in countries with either high or low TB incidence. Several systematic reviews have demonstrated its preventive efficacy. A systematic review of RCTs involving people living with HIV (17) showed that isoniazid monotherapy reduces the overall risk for TB by 33% (RR 0.67; 95% CI 0.51;0.87), and the preventive efficacy reached 64% for people with a positive TST (RR 0.36; 95% CI 0.22;0.61). Furthermore, the efficacy of the 6-month regimen was not significantly different from that of 12 months’ daily isoniazid monotherapy (RR 0.58; 95% CI 0.3;1.12) (17). A recent systematic review of RCTs also showed a significantly greater reduction in TB incidence among participants given the 6-month regimen than in those given a placebo (odds ratio, 0.65; 95% CI 0.50;0.83) (45).

9 months’ isoniazid monotherapyNo controlled clinical trials were found of 9 months’ versus 6 months’ isoniazid monotherapy. Re-analysis and modelling of the United States Public Health Service trials of isoniazid conducted in the 1950s and 1960s, however, showed that the benefit of isoniazid increases progressively when it is given for up to 9–10 months and stabilizes thereafter (46).

36 months’ isoniazid monotherapy A systematic review and meta-analysis of three RCTs of people living with HIV in settings with high TB prevalence and transmission showed that continuous IPT can reduce the risk for active TB by 38% more than 6 months’ isoniazid (47). The effect was greater in people with a positive TST (49% for active TB and 50% for death). In those with a negative TST, neither effect was significant, although the point estimate indicated a reduction in TB incidence of 27%.

5.2 Daily rifampicin plus isoniazid for 3–4 monthsA systematic review (48) and its update in 2017 (45) showed that the efficacy and the safety profile of 3–4 months’ daily rifampicin plus isoniazid were similar to those of 6 months’ isoniazid. The previous GDG therefore concluded that daily rifampicin plus isoniazid could be used as an alternative to isoniazid in settings with low TB incidence (9, 10). We conducted a new review to compare the effectiveness in children of rifampicin plus isoniazid daily for 3 months with isoniazid for 6 or 9 months. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2. We identified one RCT and two observational studies (49–51). The authors of the RCT (50) reported no clinical disease in either group and used new radiographic findings suggestive of active TB as a proxy for clinical disease. Fewer participants given daily rifampicin plus isoniazid than those given 9 months of isoniazid developed radiographic changes (RR 0.49, 95% CI 0.32;0.76). The authors also reported a lower risk for adverse events (RR 0.33, 95% CI 0.20;0.56) and a higher adherence rate (RR 1.07, 95% CI 1.01;1.14) among children given daily rifampicin plus isoniazid. Similar findings were reported in the two observational studies (49, 51).

5.3 Daily rifampicin monotherapy for 3–4 monthsA previous systematic review conducted for the 2015 LTBI guidelines (48), updated in 2017 (45), found similar efficacy for 3–4 months’ daily rifampicin and 6 months of isoniazid (odds ratio, 0.78; 95% CI, 0.41;1.46). The review also showed that individuals given rifampicin daily for 3–4 months had a lower risk for hepatotoxicity than those treated with isoniazid monotherapy (odds ratio, 0.03; 95% CI 0.00;0.48).

25

5.4 Weekly rifapentine plus isoniazid for 3 monthsA new systematic review was conducted to compare the effectiveness of a 3-month weekly regimen of rifapentine plus isoniazid with that of isoniazid monotherapy. The review covered four RCTs (52–55), which were analysed for three subgroups: adults with HIV infection, adults without HIV infection and children and adolescents, who could not be stratified according to HIV status because the relevant studies were lacking. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2.

Two of the RCTs involved adults with HIV from South Africa, Peru and in a number of countries with a low incidence of TB. No significant difference was found in the incidence of active TB between participants given a 3-month weekly regimen of rifapentine plus isoniazid and 6 or 9 months of isoniazid monotherapy (RR 0.73, 95% CI 0.23;2.30). Furthermore, the risk for hepatotoxicity was significantly lower with the 3-month weekly regimen of rifapentine plus isoniazid in adults with HIV (RR 0.26, 95% CI 0.12;0.55) and those without HIV (RR 0.16, 95% CI 0.10;0.27). The weekly regimen was also associated with a higher completion rate in all subgroups (adults with HIV: RR 1.25, 95% CI 1.01;1.55; adults without HIV: RR 1.19, 95% CI 1.16;1.22, children and adolescents: RR 1.09, 95% CI 1.03;1.15). One RCT included a comparison between a 3-month weekly regimen of rifapentine plus isoniazid and continuous isoniazid monotherapy in adults with HIV infection (52). No significant difference in TB incidence was found in an intention-to-treat analysis; however, a per-protocol analysis showed a lower rate of TB infection or death in participants given continuous isoniazid. In all the studies, the 3-month regimen of weekly rifapentine plus isoniazid was given under direct observation.

In a study of the 3-month weekly regimen of rifapentine plus isoniazid in 112 pregnant women, the rates of spontaneous abortion and birth defects were similar to those in the general US population (56).

Table 2 Recommended dosages of drugs for the treatment of LTBI

Drug regimen Dose per kg body weight Maximum dose

Isoniazid alone, daily for 6 or 9 months Adults, 5 mgChildren, 10 mg (range, 7–15 mg) 300 mg

Daily rifampicin alone for 3–4 months Adults, 10 mgChildren, 15 mg (range, 10–20 mg) 600 mg

Daily isoniazid plus rifampicin for 3–4 months

Isoniazid:Adults, 5 mgChildren, 10 mg (range, 7–15 mg)RifampicinAdults, 10 mgChildren, 15 mg (range, 10–20 mg)

Isoniazid, 300 mgRifampicin, 600 mg

Weekly rifapentine plus isoniazid for 3 months (12 doses)

Individuals aged ≥ 12 years: Isoniazid: 15 mgIndividuals aged 2–11 years: isoniazid: 25 mgRifapentine:10.0–14.0 kg = 300 mg14.1–25.0 kg = 450 mg25.1–32.0 kg = 600 mg32.1–50.0 kg = 750 mg> 50 kg = 900 mg

Isoniazid, 900 mgRifapentine, 900 mg

Rationale for the recommendations The selection of treatment options by programmes and clinicians should consider the characteristics of the clients who are to receive treatment to ensure that it is not only initiated but also completed. The GDG agreed that the benefits of all the treatment options outweigh the potential harm. They also noted that all the treatment options can be self-administered. An RCT showed that self-administered treatment of the 3-month regimen of weekly rifapentine plus isoniazid is not inferior to directly observed treatment (57); however, there is little further evidence on self-administration of this regimen. The GDG noted that a requirement for a direct observation could be a significant barrier to the implementation. The GDG further noted that individuals receiving treatment, clinicians providing treatment and programme managers prefer shorter to longer regimens.



On the basis of existing practice, albeit, in the absence of a direct comparison, the GDG judged that 9 months’ isoniazid is an equivalent option to 6 months of isoniazid in countries with a low TB incidence and a strong health infrastructure. It noted, however, that 6 months’ isoniazid is preferable to 9 months from the point of view of feasibility, resource requirements and acceptability to patients.

The recommendation that people living with HIV should be given at least 36 months of daily isoniazid monotherapy is conditional, as continuous IPT depends on the epidemiology and transmission of TB, the health infrastructure and programme priorities.

The GDG agreed unanimously that the benefits of 3 months’ daily rifampicin plus isoniazid for infants and children < 15 years of age outweigh the harm, given its safety profile, the higher rate of completion as compared with isoniazid monotherapy and the availability of child-friendly, fixed-dose combinations of rifampicin and isoniazid. The GDG therefore made a strong recommendation despite the low quality of the evidence.

Considerations for implementation Drug resistance and surveillance There is no evidence of a significant association between bacterial resistance to TB drugs and use of isoniazid or rifamycins for the treatment of LTBI (58, 59). Nonetheless, active TB disease must be excluded before TB preventive treatment is initiated (section 3), and regular follow-up is required to ensure early identification of people who develop active TB while receiving TB preventive treatment. National surveillance systems for resistance to TB drugs should be established in countries implementing programmatic management of LTBI.

Interactions with antiretroviral drugs Regimens containing rifampicin and rifapentine should be prescribed with caution to people living with HIV who are on ART because of potential drug–drug interactions. These regimens should not be administered to people receiving protease inhibitors or nevirapine. The GDG noted that the 3-month regimen of weekly rifapentine plus isoniazid can be administered to patients receiving efavirenz-based antiretroviral regimens without dose adjustment, according to a study of pharmacokinetics (60). Administration of rifapentine with raltegravir was found to be safe and well tolerated (61). Rifapentine-containing regimens should not be administered with dolutegravir until more information becomes available. The GDG stressed the urgent need for studies of the pharmacokinetics of the 3-month regimen of weekly rifapentine plus isoniazid concomitantly with other drugs, particularly ART.

Monitoring of adverse events As individuals who receive treatment for LTBI do not have active disease, their risk for adverse events during treatment must be minimized. Adverse reactions have been associated with isoniazid (asymptomatic elevation of serum liver enzyme concentrations, peripheral neuropathy and hepatotoxicity), rifampicin and rifapentine (cutaneous reactions, hypersensitivity reactions, gastrointestinal intolerance and hepatotoxicity). While most of these reactions are minor and occur rarely, specific attention should be paid to preventing drug-induced hepatotoxicity.

Individuals receiving treatment for LTBI should be monitored routinely at monthly visits to health care providers, who should explain the disease process and the rationale of the treatment and emphasize the importance of completing it. Patients receiving treatment should be advised to contact their health care provider at any time if they become aware of symptoms such as anorexia, nausea, vomiting, abdominal discomfort, persistent fatigue or weakness, dark-coloured urine, pale stools or jaundice. If a health care provider cannot be consulted at the onset of such symptoms, the patient should stop treatment immediately.

There is insufficient evidence to support testing of baseline liver function (62). It is, however, strongly encouraged, where feasible, for individuals with the following risk factors: history of liver disease, regular use of alcohol, chronic liver disease, HIV infection, age > 35 years, pregnancy or in the immediate postpartum period (within 3 months of delivery). For individuals with abnormal baseline test results, sound clinical judgement is required to ensure that the benefit of TB preventive treatment outweighs the risks, and they should be tested

27

routinely at subsequent visits. Appropriate laboratory testing should also be performed for patients who become symptomatic while on treatment (e.g. liver function tests for those with symptoms of hepatotoxicity).

Individuals at risk for peripheral neuropathy, such as those with malnutrition, chronic alcohol dependence, HIV infection, renal failure or diabetes, or who are pregnant or breastfeeding, should receive vitamin B6 supplements when taking isoniazid-containing regimens.

Adherence to and completion of preventive treatmentAdherence to the full course and completion of treatment are important determinants of clinical benefit to individuals and to the success of programmes. A systematic review conducted for the WHO 2015 LTBI guidelines (9, 10) provided heterogeneous results for interventions to improve treatment adherence and completion, and the evidence was considered inconclusive. The GDG noted that interventions to ensure adherence and completion of treatment should be tailored to the specific needs of risk groups and the local context. The WHO guidelines for treatment of drug-susceptible active TB propose several interventions to support adherence (63), which could be applied to treatment of LTBI. Fixed-dose combinations such as co-trimoxazole plus isoniazid and isoniazid plus rifampicin should be used where possible to reduce the number of pills to be taken. Concerns about adherence should not be a barrier to use of preventive treatment.



6. Preventive treatment for contacts of patients with multidrug-resistant tuberculosis

In selected high-risk household contacts of patients with multidrug-resistant tuberculosis, preventive treatment may be considered based on individualised risk assessment and a sound clinical justification. (Conditional recommendation, very low-quality evidence. New recommendation)

Remarks

The preventive treatment should be individualized after a careful assessment of the intensity of exposure, the certainty of the source case, reliable information on the drug resistance pattern of the source case and potential adverse events.

The preventive treatment should be given only to household contacts at high risk (e.g. children, people receiving immunosuppressive therapy and people living with HIV).

The drugs should be selected according to the drug susceptibility profile of the source case.

Confirmation of infection with LTBI tests is required.

This recommendation must not affect on-going placebo-controlled clinical trials of MDR-TB contacts on ethical grounds. The results of such clinical trials are crucial for updating this recommendation.

Strict clinical observation and close monitoring for the development of active TB disease for at least 2 years are required, regardless of the provision of preventive treatment.

Summary of evidence A systematic review of the effectiveness of preventive treatment for contacts of people with MDR-TB conducted for the 2015 LTBI guidelines (9, 10) was updated. The updated review comprised 10 studies (6 newly identified and 4 from the previous review) that allowed comparisons between participants who received preventive treatment for MDR-TB and those who did not. Because of clinical heterogeneity among the studies, a meta-analysis could not be performed. Of the 10 studies, one was excluded because only isoniazid monotherapy was used, and an additional five studies were excluded as fewer than 20 participants completed preventive TB treatment. Therefore, the quality of the evidence was based on only four studies. The evidence-to-decision and the GRADE tables are presented in Annexes 1 and 2.

No TB case was reported in either the intervention or the control group in one study (64), while one case of active TB due to a drug-susceptible strain that was different from that of the source case was reported in another study (65). The remaining two studies (66, 67) addressed the efficacy of preventive treatment. In one study of 119 cohorts, 104 contacts with LTBI initiated fluoroquinolone-based preventive treatment, of whom 93 (89%) completed treatment, and none developed active TB; while 3 of 15 (20%) contacts who refused treatment developed MDR-TB (odds ratio, 0.02, 95% CI 0.00;0.39) (66). In the other study, confirmed or probable TB developed in 2 of 41 (4.9%) children receiving tailored preventive treatment and in 13 of 64 (20.3%) children who did not receive proper preventive treatment (odds ratio, 0.2, 95% CI 0.04;0.94) (67).

The drugs used in these studies were mainly fluoroquinolones (e.g. moxifloxacin, levofloxacin) with or without other agents (e.g. ethambutol, ethionamide). None of studies included a comparison of the risk for adverse events, although one reported that no serious adverse events could be attributed to fluoroquinolone-based preventive treatment (66). The median proportion of participants who discontinued treatment because of adverse events in all the studies was 5.1% (interquartile range, 1.9–30.2%).

29

Rationale for the recommendations Overall, the GDG judged that the potential benefits of targeted preventive treatment for MDR-TB contacts based on individual risk assessments outweigh the harm but acknowledged uncertainty about the efficacy of the intervention due to the lack of RCTs. It also noted that provision of preventive treatment for MDR-TB contacts would be acceptable, particularly to patients and health care workers. The GDG stressed that treatment should be given to selected individuals after a careful risk assessment, including intensity of exposure, certainty of the source case, reliable information on the drug resistance pattern of the index case and potential adverse events. It should be given only to household contacts at high risk (e.g. children, people on immunosuppressive therapy and people living with HIV). Confirmation of infection by LTBI testing is required before individualized treatment is initiated.

Considerations for implementation Close monitoring and treatment adherenceClose monitoring of adverse events and adherence to treatment is essential. The types of adverse events depend on the drugs used. Common adverse events associated with each drug are listed in the Companion handbook to the WHO guidelines for the programmatic management of drug-resistant tuberculosis (68). Adverse effects should be monitored according to the WHO framework for monitoring and managing the safety of drugs against active TB (69). The GDG reiterated that strict clinical observation and close monitoring for active TB disease based on sound clinical practice and national guidelines for at least 2 years is required, regardless of the provision of preventive treatment. Consideration should also be given to interactions with antiretroviral, immunosuppressant and other drugs when providing TB preventive treatment.

Informed consent As the recommendation is based on very low-quality evidence, clients must be given detailed information about the benefits and harms of the preventive treatment and asked for explicit informed consent. In view of the uncertainty about the balance of benefit to harm, informed consent, preferably in writing, is required, based on the local context and practice in similar situations.

Selection of drug regimenThe regimen of preventive treatment of MDR-TB contacts should be based on reliable information on the drug resistance profile of the source case. Later-generation fluoroquinolones (e.g. levofloxacin and moxifloxacin) are considered to be important components of a preventive treatment regimen unless the strain of the source case is resistant to them. Although there has been concern about the use of fluoroquinolones in children because retardation of cartilage development was shown in animals (70), similar effects have not been demonstrated in humans (68, 71, 72). There is limited evidence for the duration of treatment, and this should be based on clinical judgement. The regimens used in the studies conducted so far were given for 6, 9 and 12 months.

Resources and feasibilityFor a programmatic approach, all the necessary resources should be in place, including for quality-assured testing for drug susceptibility, the necessary medications and a system for close monitoring of harm and adverse events. The feasibility of providing preventive treatment should be carefully assessed according to the availability of resources and the history and status of preventive treatment for drug-susceptible TB.

6. PREVENTIVE TREATMENT FOR CONTACTS OF PEOPLE WITH MULTIDRUG-RESISTANT TUBERCULOSIS


7. Issues in implementation

7.1 Ethical considerationsLTBI testing and treatment raise a range of ethical issues (73, 74). First, LTBI is by definition an asymptomatic state. This alters the ethical obligations that would be imposed by active TB. For example, the absence of an immediate risk of transmission makes it unethical to restrict migration on the basis of the LTBI status of an individual. Secondly, the difficulty of accurate assessment of individual risk for the development of active TB poses a challenge to communication. Informed consent requires effective, adequate communication of the uncertainty in LTBI testing, the risk for development of active TB, possible side-effects of treatment and the protective benefits. Risk and uncertainty must be communicated in culturally and linguistically appropriate forms and feedback obtained after screening programmes. Thirdly, LTBI disproportionately affects individuals and groups that are already socially and medically vulnerable. Therefore, efforts must be made to ensure equity and human rights, so that the vulnerability of target groups does not preclude their access to screening and treatment. Any intervention for vulnerable groups should include minimization of the risk for stigmatization. Policies should be evaluated from an ethical perspective after implementation, both to consider possible unexpected effects and to ensure that the evidence on which they are based remains current and relevant (75).

7.2 Programme management, monitoring and evaluationNational programmes should prepare a national plan for programmatic LTBI management, including prioritization of groups at high risk on the basis of local epidemiology and the characteristics of the health system. They should create a conducive policy and programmatic environment, including national and local policies and standard operating procedures to facilitate implementation of the recommendations in these guidelines. This should include promoting universal health coverage and offering public financing for LTBI management. Furthermore, dedicated resources should be allocated, including for human resource development and service delivery in the community. National TB programmes should ensure meaningful engagement with affected populations, their communities, the private sector, other relevant health programmes and line ministries in both planning and implementing the interventions. The national TB programme is also encouraged to ensure access to comprehensive care for co-existing risk factors for TB, such as diabetes, undernutrition and tobacco smoking.

Preventive TB treatment for people living with HIV should be a core component of HIV preventive care and should be the responsibility of national AIDS programmes and HIV service providers. Preventive treatment should not be viewed as an isolated intervention but should be part of a comprehensive package of care.

Coverage of contact investigation and treatment of LTBI among child contacts and people living with HIV are among the top 10 core indicators for monitoring implementation of the End TB Strategy. Programmatic management of LTBI should include monitoring and evaluation systems that are aligned with national patient monitoring and surveillance systems (76). Appropriate recording and reporting tools should be developed, and standardized indicators (Table 3) should be measured to regularly inform decision-making for programme implementation. Some may require changes to national legislation (e.g. making LTBI a notifiable condition) and to the policy framework, which should be addressed according to the local and national context. It is important to engage the private health sector and to ensure proper recording and reporting from both the private and public sectors. Use of electronic case-based monitoring will facilitate recording and reporting for LTBI management.

31

Tabl

e 3

Mon

itori

ng a

nd e

valu

atio

n in

dica

tors

for p

rogr

amm

atic

man

agem

ent o

f LTB

I rec

omm

ende

d by

WH

O

IND

ICAT

OR

NU

MER

ATO

RD

ENO

MIN

ATO

RPU

RPO

SE

CORE

GLO

BAL

AN

D N

ATIO

NA

L IN

DIC

ATO

RS

1Pr

opor

tion

of c

hild

ren

< 5

year

s w

ho a

re h

ouse

hold

co

ntac

ts o

f TB

case

s (a

ccor

ding

to n

atio

nal

guid

elin

es) w

ho h

ave

com

plet

ed in

vest

igat

ions

fo

r TB

Tota

l num

ber o

f chi

ldre

n <

5 ye

ars

who

are

ho

useh

old

cont

acts

of T

B ca

ses

who

com

plet

ed

inve

stig

atio

ns fo

r TB

(acc

ordi

ng to

nat

iona

l gu

idel

ines

) dur

ing

the

repo

rtin

g pe

riod

Tota

l num

ber o

f chi

ldre

n <

5 ye

ars

who

wer

e ho

useh

old

cont

acts

of T

B ca

ses

(acc

ordi

ng to

na

tiona

l gui

delin

es) d

urin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to

ensu

re e

ffect

ive

cont

act i

nves

tigat

ion,

a k

ey

com

pone

nt o

f TB

diag

nosi

s an

d pr

even

tion

amon

g ch

ildre

n <

5 ye

ars

2Pr

opor

tion

of c

hild

ren

< 5

year

s ol

d w

ho a

re

hous

ehol

d co

ntac

ts o

f TB

case

s (a

ccor

ding

to

nat

iona

l gui

delin

es) w

ho a

re e

ligib

le fo

r TB

prev

entiv

e tr

eatm

ent w

ho h

ave

star

ted

trea

tmen

t

Tota

l num

ber o

f chi

ldre

n <

5 ye

ars

who

are

ho

useh

old

cont

acts

of T

B ca

ses

who

sta

rted

TB

prev

entiv

e tr

eatm

ent d

urin

g th

e re

port

ing

perio

d

Tota

l num

ber o

f chi

ldre

n <

5 ye

ars

who

are

ho

useh

old

cont

acts

of T

B ca

ses

who

wer

e el

igib

le fo

r TB

prev

entiv

e tr

eatm

ent d

urin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to

initi

ate

trea

tmen

t in

child

ren

< 5

year

s w

ho

are

hous

ehol

d co

ntac

ts a

nd a

re e

ligib

le fo

r TB

prev

entiv

e tr

eatm

ent

3Pr

opor

tion

of e

ligib

le p

eopl

e liv

ing

with

HIV

, new

ly

enro

lled

in H

IV c

are

and

star

ted

on T

B pr

even

tive

trea

tmen

t

Tota

l num

ber o

f elig

ible

peo

ple

livin

g w

ith H

IV

who

sta

rted

TB

prev

entiv

e tr

eatm

ent d

urin

g th

e re

port

ing

perio

d

Tota

l num

ber o

f elig

ible

peo

ple

livin

g w

ith

HIV

and

new

ly e

nrol

led

in H

IV c

are

durin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to

initi

ate

trea

tmen

t of a

ll in

divi

dual

s in

HIV

car

e w

ho a

re e

ligib

le fo

r TB

prev

entiv

e tr

eatm

ent

CORE

NAT

ION

AL

IND

ICAT

ORS

4Pr

opor

tion

of e

ligib

le in

divi

dual

s in

at-

risk

popu

latio

ns (a

s de

fined

by

natio

nal g

uide

lines

) te

sted

for L

TBI i

nfec

tion

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

w

ho w

ere

test

ed fo

r LTB

I dur

ing

the

repo

rtin

g pe

riod

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

who

wer

e el

igib

le fo

r tes

ting

durin

g th

e re

port

ing

perio

d

Mea

sure

s co

vera

ge o

f tes

ting

at-r

isk

popu

latio

ns e

ligib

le fo

r TB

prev

entiv

e tr

eatm

ent

5Pr

opor

tion

of in

divi

dual

s in

at-

risk

popu

latio

ns

(acc

ordi

ng to

nat

iona

l gui

delin

es) w

ith a

pos

itive

LT

BI te

st w

ho a

re e

ligib

le fo

r TB

prev

entiv

e tr

eatm

ent a

nd w

ho h

ave

star

ted

trea

tmen

t

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

w

ho s

tart

ed T

B pr

even

tive

trea

tmen

t dur

ing

the

repo

rtin

g pe

riod

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

who

hav

e te

sted

pos

itive

for L

TBI

and

wer

e el

igib

le fo

r TB

prev

entiv

e tr

eatm

ent

durin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to in

itiat

e tr

eatm

ent o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

who

are

elig

ible

for T

B pr

even

tive

trea

tmen

t

6Pr

opor

tion

of e

ligib

le in

divi

dual

s in

at-

risk

popu

latio

ns (a

ccor

ding

to n

atio

nal g

uide

lines

) w

ith a

pos

itive

LTBI

test

who

sta

rted

TB

prev

entiv

e tr

eatm

ent a

nd c

ompl

eted

the

cour

se

Num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

who

co

mpl

eted

the

cour

se o

f TB

prev

entiv

e tr

eatm

ent

durin

g th

e re

port

ing

perio

d

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

with

a p

ositi

ve LT

BI te

st w

ho

star

ted

TB p

reve

ntiv

e tr

eatm

ent d

urin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to

ensu

re th

at in

divi

dual

s in

at-

risk

popu

latio

ns

adhe

re to

the

full

cour

se o

f tre

atm

ent

7Pr

opor

tion

of e

ligib

le p

eopl

e liv

ing

with

HIV

who

co

mpl

eted

a c

ours

e of

TB

prev

entiv

e tr

eatm

ent

Num

ber o

f elig

ible

peo

ple

livin

g w

ith H

IV w

ho

com

plet

ed a

cou

rse

of T

B pr

even

tive

trea

tmen

t du

ring

the

repo

rtin

g pe

riod

Tota

l num

ber o

f elig

ible

peo

ple

livin

g w

ith H

IV

who

sta

rted

TB

prev

entiv

e tr

eatm

ent d

urin

g th

e re

port

ing

perio

d

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to

ensu

re th

at p

eopl

e liv

ing

with

HIV

adh

ere

to

the

full

cour

se o

f tre

atm

ent

8Pr

opor

tion

of c

hild

ren

< 5

year

s w

ho a

re h

ouse

hold

co

ntac

ts o

f TB

case

s (a

ccor

ding

to n

atio

nal

guid

elin

es) w

ho h

ave

com

plet

ed a

cou

rse

of T

B pr

even

tive

trea

tmen

t

Num

ber o

f chi

ldre

n <

5 ye

ars

who

are

hou

seho

ld

cont

acts

of T

B ca

ses

(acc

ordi

ng to

nat

iona

l gu

idel

ines

) who

com

plet

ed a

cou

rse

of T

B pr

even

tive

trea

tmen

t dur

ing

the

repo

rtin

g pe

riod

Tota

l num

ber o

f chi

ldre

n <

5 ye

ars

who

are

ho

useh

old

cont

acts

of T

B ca

ses

(acc

ordi

ng to

na

tiona

l gui

delin

es) w

ho s

tart

ed T

B pr

even

tive

trea

tmen

t dur

ing

the

repo

rtin

g pe

riod

Mea

sure

s th

e ca

paci

ty o

f the

pro

gram

me

to e

nsur

e th

at c

hild

ren

< 5

year

s w

ho a

re

hous

ehol

d co

ntac

ts o

f TB

case

s ad

here

to th

e fu

ll co

urse

of t

reat

men

t

OPT

ION

AL

IND

ICAT

OR

9TB

inci

denc

e ra

te in

at-

risk

popu

latio

ns (a

s de

fined

by

nat

iona

l gui

delin

es)

Tota

l num

ber o

f new

ly n

otifi

ed T

B ca

ses

in a

t-ris

k po

pula

tions

dur

ing

the

repo

rtin

g pe

riod

Tota

l num

ber o

f ind

ivid

uals

in a

t-ris

k po

pula

tions

Mea

sure

s th

e im

pact

of t

he p

rogr

amm

e on

the

inci

denc

e of

TB

in a

t-ris

k po

pula

tions

From

refe

renc

e (1

2)

7. ISSUES IN IMPLEMENTATION


8. Research priorities

The review of the evidence for the recommendations exposed gaps in research that should be filled urgently, in addition to those identified during preparation of the 2015 guidelines.

8.1 Risks of at-risk populations for progression to active TB disease Evidence on the risks of a number of at-risk populations for progression from LTBI to active disease will be crucial for determining the potential benefits of LTBI treatment and for designing appropriate public health interventions. In particular, strong evidence from clinical trials is lacking for the following groups: patients with diabetes, people with harmful use of alcohol, tobacco smokers, underweight people, people exposed to silica, patients receiving steroid treatment, patients with rheumatological conditions, indigenous populations and cancer patients. Both direct measurement of the incidence of active TB and methods for measuring the risk for active TB disease could be explored, such as use of genotyping to measure the risk for reactivation.

Evidence is also required on differential harm and the acceptability of testing and treatment for LTBI in specific risk groups, including socially adverse events such as stigmatization.

8.2 Defining the best algorithm for ruling out active TBOperational and clinical studies should be conducted to exclude active TB before preventive treatment is given. The performance and feasibility of the algorithms proposed in these guidelines should be assessed. In particular, few data are available on children and pregnant women. Strategies to save cost and improve feasibility (e.g. use of mobile chest radiography) should also be explored.

8.3 Improved diagnostic tests and performance of LTBI tests in at-risk populations

Diagnostic tests with improved performance and predictive value for reactivation of TB are critically needed. In addition, the performance of LTBI tests should be evaluated in various at-risk populations, such as the best way of using the available tools (e.g. combination or sequential use of TST and IGRA) in each at-risk population.

8.4 Treatment options for LTBI Research to find shorter, better-tolerated treatment regimens than those currently recommended is a priority. Studies of efficacy and adverse events in certain risk groups (e.g. people who use drugs, people with alcohol use disorder and elderly people) are essential. In particular, there are no or very limited data on the use of rifapentine in children < 2 years and pregnant women. Studies should be conducted of the pharmacokinetics of interactions between rifamycin-containing regimens and other drugs, particularly antiretroviral drugs. In addition, the durability of protection by preventive treatment should be evaluated in settings in which TB is endemic, including the efficacy of repeated courses of preventive treatment.

8.5 Monitoring of adverse eventsProspective randomized studies are required to determine the incremental benefits of routine monitoring of liver enzyme levels over education and clinical observation alone for preventing severe clinical adverse events, with stratification of the evidence by at-risk population.

338. RESEARCH PRIORITIES

8.6 Risk of drug resistance following LTBI treatment Programme-based surveillance systems and clinical studies are needed to monitor the risk for bacterial resistance to the drugs used in LTBI treatment. Particular consideration should be given to rifamycin-containing regimens because of the dearth of data.

8.7 Adherence to and completion of treatmentCarefully designed studies, including RCTs, are required to generate evidence on the effectiveness of context-specific interventions for enhancing adherence and completion of treatment. The studies should include specific risk groups, depending on the available resources and the health system infrastructure. Use of “digital health” to improve adherence is an important area. Further research is required on the effectiveness of self-administration of the 3-month regimen of weekly rifapentine plus isoniazid.

8.8 Cost–effectiveness Although a number of studies of the cost-effectiveness of TB preventive treatment are available, their wide heterogeneity obviates a comprehensive appraisal of the cost-effectiveness of LTBI management stratified by population group and type of intervention. Direct measurement of cost-effectiveness in certain settings and populations would allow extension of the LTBI strategy at national or local level.

8.9 Preventive treatment for contacts of people with MDR-TB RCTs with adequate power are urgently needed to update the recommendation on preventive treatment for contacts of people with MDR-TB. Trials should be performed with both adult and paediatric populations and with at-risk populations such as people living with HIV. The composition, dosage and duration of preventive treatment regimens for MDR-TB should be optimized, and the potential role of newer drugs with good sterilization properties should be investigated. The effectiveness and safety of preventive treatment for contacts of people with MDR-TB should be evaluated in operational conditions. Further evidence on the risk of contacts of people with MDR-TB for progression to active TB will be important for understanding the benefits of preventive treatment.

8.10 Programme management Epidemiological research should be conducted to determine the burden of LTBI in various geographical settings and risk groups and as a basis for nationally and locally tailored interventions, including integrated community-based approaches. Research is also needed on service delivery models to ensure that patients are properly managed including the provision of additional interventions for tobacco smokers, illicit drug users, and people with harmful use of alcohol. Household implementation models could improve the effectiveness and efficiency of delivery of interventions. Tools should be developed and assessed to facilitate monitoring and evaluation of programmatic management of LTBI.


9. References

1. Getahun H, Matteelli A, Chaisson RE, Raviglione M. Latent Mycobacterium tuberculosis infection. N Engl J Med. 2015;372(22):2127–35.

2. Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, et al. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med. 2003;163(9):1009–21.

3. Dodd PJ, Gardiner E, Coghlan R, Seddon JA. Burden of childhood tuberculosis in 22 high-burden countries: a mathematical modelling study. Lancet Glob Health. 2014;2(8):e453–9.

4. Houben RM, Dodd PJ. The Global Burden of Latent Tuberculosis Infection: A Re-estimation Using Mathematical Modelling. PLoS Med. 2016;13(10):e1002152.

5. Comstock GW, Livesay VT, Woolpert SF. The prognosis of a positive tuberculin reaction in childhood and adolescence. Am J Epidemiol. 1974;99(2):131–8.

6. Uplekar M, Weil D, Lonnroth K, Jaramillo E, Lienhardt C, Dias HM, et al. WHO’s new end TB strategy. Lancet. 2015;385(9979):1799–801.

7. WHO. Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with HIV in resource-constrained settings. 2011.

8. WHO. Guidance for National tuberculosis programmes on the management of tuberculosis in children. Second edition. 2014.

9. WHO. Guidelines on the management of latent tuberculosis infection. 2015.

10. Getahun H, Matteelli A, Abubakar I, Aziz MA, Baddeley A, Barreira D, et al. Management of latent Mycobacterium tuberculosis infection: WHO guidelines for low tuberculosis burden countries. Eur Respir J. 2015;46(6):1563–76.

11. WHO. Recommendations for Investigating Contacts of Persons with Infectious Tuberculosis in Low- and Middle-Income Countries. 2012.

12. WHO. Global TB Report. 2017.

13. 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(4):928–52.

14. WHO. WHO Handbook for Guideline Development – 2nd Edition. 2014.

15. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924.

16. Ford N, Matteelli A, Shubber Z, Hermans S, Meintjes G, Grinsztejn B, et al. TB as a cause of hospitalization and in-hospital mortality among people living with HIV worldwide: a systematic review and meta-analysis. J Int AIDS Soc. 2016;19(1):20714.

17. Akolo C, Adetifa I, Shepperd S, Volmink J. Treatment of latent tuberculosis infection in HIV infected persons. Cochrane Database Syst Rev. 2010;1.

18. Rangaka MX, Wilkinson RJ, Boulle A, Glynn JR, Fielding K, van Cutsem G, et al. Isoniazid plus antiretroviral therapy to prevent tuberculosis: a randomised double-blind, placebo-controlled trial. Lancet. 2014.

359. REFERENCES

19. The TEMPRANO ANRS 12136 Study Group. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. N Engl J Med. 2015;373(9):808–22.

20. Badje A, Moh R, Gabillard D, Guehi C, Kabran M, Ntakpe JB, et al. Effect of isoniazid preventive therapy on risk of death in west African, HIV-infected adults with high CD4 cell counts: long-term follow-up of the Temprano ANRS 12136 trial. Lancet Glob Health. 2017;5(11):e1080-e9.

21. Bruins WS, van Leth F. Effect of secondary preventive therapy on recurrence of tuberculosis in HIV-infected individuals: a systematic review. Infect Dis (Lond). 2017;49(3):161–9.

22. Gupta A, Nayak U, Ram M, Bhosale R, Patil S, Basavraj A, et al. Postpartum Tuberculosis Incidence and Mortality among HIV-Infected Women and Their Infants in Pune, India, 2002–2005. Clinical Infectious Diseases. 2007;45(2):241–9.

23. Bothamley G. Drug Treatment for Tuberculosis during Pregnancy: Safety Considerations. Drug Safety. 2001;24(7):553–65.

24. WHO. Recommendation on 36 months isoniazid preventive therapy to adults and adolescents living with HIV in resource-constrained and high TB and HIV-prevalence settings: 2015 update. 2015.

25. Zar HJ, Cotton MF, Strauss S, Karpakis J, Hussey G, Schaaf HS, et al. Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: randomised controlled trial. BMJ. 2007;334(7585):136.

26. Madhi SA, Nachman S, Violari A, Kim S, Cotton MF, Bobat R, et al. Primary Isoniazid Prophylaxis against Tuberculosis in HIV-Exposed Children. New England Journal of Medicine. 2011;365(1):21–31.

27. Gray DM, Workman LJ, Lombard CJ, Jennings T, Innes S, Grobbelaar CJ, et al. Isoniazid preventive therapy in HIV-infected children on antiretroviral therapy: a pilot study. Int J Tuberc Lung Dis. 2014;18(3):322–7.

28. Frigati LJ, Kranzer K, Cotton MF, Schaaf HS, Lombard CJ, Zar HJ. The impact of isoniazid preventive therapy and antiretroviral therapy on tuberculosis in children infected with HIV in a high tuberculosis incidence setting. Thorax. 2011;66(6):496–501.

29. WHO. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection Recommendations for a public health approach – Second edition. 2016.

30. Getahun H, Kittikraisak W, Heilig CM, Corbett EL, Ayles H, Cain KP, et al. Development of a standardized screening rule for tuberculosis in people living with HIV in resource-constrained settings: individual participant data meta-analysis of observational studies. PLoS Med. 2011;8(1):e1000391.

31. WHO. Chest radiography in tuberculosis detection-Summary of current WHO recommendations and guidance on programmatic approaches. 2016.

32. Cranmer. L, Pavlinac. P, Njuguna. I, Otieno. V, Maleche-Obimbo. E, Moraa. H, et al. Performance of WHO TB symptom screen in hospitalized HIV-positive Kenyan children. 47th World Conference on Lung Health of the International Union Against Tuberculosis and Lung Disease; 2016; Liverpool, United Kingdom.

33. van’t Hoog AHL, Mitchell E, et al. A systematic review of the sensitivity and specificity of symptom- and chestradiography screening for active pulmonary tuberculosis in HIVnegative persons and persons with unknown HIV status. Report to WHO. 2013. www.who.int/tb/Review2Accuracyofscreeningtests.pdf. 2013.

34. Mulenga H, Tameris MD, Luabeya KK, Geldenhuys H, Scriba TJ, Hussey GD, et al. The Role of Clinical Symptoms in the Diagnosis of Intrathoracic Tuberculosis in Young Children. Pediatr Infect Dis J. 2015;34(11):1157–62.

35. Kruk A, Gie RP, Schaaf HS, Marais BJ. Symptom-based screening of child tuberculosis contacts: improved feasibility in resource-limited settings. Pediatrics. 2008;121(6):e1646–52.


36. Triasih R, Robertson CF, Duke T, Graham SM. A prospective evaluation of the symptom-based screening approach to the management of children who are contacts of tuberculosis cases. Clin Infect Dis. 2015;60(1):12–8.

37. Mandalakas AM, Hesseling AC, Gie RP, Schaaf HS, Marais BJ, Sinanovic E. Modelling the cost-effectiveness of strategies to prevent tuberculosis in child contacts in a high-burden setting. Thorax. 2013;68(3):247–55.

38. Rangaka MX, Wilkinson KA, Glynn JR, Ling D, Menzies D, Mwansa-Kambafwile J, et al. Predictive value of interferon-gamma release assays for incident active tuberculosis: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12(1):45–55.

39. Mahomed H, Hawkridge T, Verver S, Abrahams D, Geiter L, Hatherill M, et al. The tuberculin skin test versus QuantiFERON TB Gold(R) in predicting tuberculosis disease in an adolescent cohort study in South Africa. PLoS One. 2011;6(3):e17984.

40. Mathad JS, Bhosale R, Balasubramanian U, Kanade S, Mave V, Suryavanshi N, et al. Quantitative IFN-gamma and IL-2 Response Associated with Latent Tuberculosis Test Discordance in HIV-infected Pregnant Women. Am J Respir Crit Care Med. 2016;193(12):1421–8.

41. McCarthy KM, Scott LE, Gous N, Tellie M, Venter WD, Stevens WS, et al. High incidence of latent tuberculous infection among South African health workers: an urgent call for action. Int J Tuberc Lung Dis. 2015;19(6):647–53.

42. Sharma SK, Vashishtha R, Chauhan LS, Sreenivas V, Seth D. Comparison of TST and IGRA in Diagnosis of Latent Tuberculosis Infection in a High TB-Burden Setting. PLoS One. 2017;12(1):e0169539.

43. Dorman SE, Belknap R, Graviss EA, Reves R, Schluger N, Weinfurter P, et al. Interferon-gamma release assays and tuberculin skin testing for diagnosis of latent tuberculosis infection in healthcare workers in the United States. Am J Respir Crit Care Med. 2014;189(1):77–87.

44. Zwerling A, Behr MA, Verma A, Brewer TF, Menzies D, Pai M. The BCG World Atlas: a database of global BCG vaccination policies and practices. PLoS Med. 2011;8(3):e1001012.

45. Zenner D, Beer N, Harris RJ, Lipman MC, Stagg HR, van der Werf MJ. Treatment of latent tuberculosis infection: An updated network meta-analysis. Annals of Internal Medicine. 2017;167(4):248–55.

46. Comstock GW. How much isoniazid is needed for prevention of tuberculosis among immunocompetent adults? Int J Tuberc Lung Dis. 1999;3(10):847–50.

47. Den Boon S, Matteelli A, Ford N, Getahun H. Continuous isoniazid for the treatment of latent tuberculosis infection in people living with HIV. AIDS. 2016;30(5):797–801.

48. Stagg HR, Zenner D, Harris RJ, Munoz L, Lipman MC, Abubakar I. Treatment of Latent Tuberculosis Infection: A Network Meta-analysis. Ann Intern Med. 2014;161(6):419–28.

49. Galli L, Lancella L, Tersigni C, Venturini E, Chiappini E, Bergamini BM, et al. Pediatric Tuberculosis in Italian Children: Epidemiological and Clinical Data from the Italian Register of Pediatric Tuberculosis. Int J Mol Sci. 2016;17(6).

50. Spyridis NP, Spyridis PG, Gelesme A, Sypsa V, Valianatou M, Metsou F, et al. The effectiveness of a 9-month regimen of isoniazid alone versus 3- and 4-month regimens of isoniazid plus rifampin for treatment of latent tuberculosis infection in children: results of an 11-year randomized study. Clin Infect Dis. 2007;45(6):715–22.

51. van Zyl S, Marais BJ, Hesseling AC, Gie RP, Beyers N, Schaaf HS. Adherence to anti-tuberculosis chemoprophylaxis and treatment in children. Int J Tuberc Lung Dis. 2006;10(1):13–8.

52. Martinson NA, Barnes GL, Moulton LH, Msandiwa R, Hausler H, Ram M, et al. New regimens to prevent tuberculosis in adults with HIV infection. N Engl J Med. 2011;365(1):11–20.

37

53. Sterling TR, Scott NA, Miro JM, Calvet G, La Rosa A, Infante R, et al. Three months of weekly rifapentine and isoniazid for treatment of Mycobacterium tuberculosis infection in HIV-coinfected persons. AIDS. 2016;30(10):1607–15.

54. Sterling TR, Villarino ME, Borisov AS, Shang N, Gordin F, Bliven-Sizemore E, et al. Three Months of Rifapentine and Isoniazid for Latent Tuberculosis Infection. New England Journal of Medicine. 2011;365(23):2155–66.

55. Villarino ME, Scott NA, Weis SE, Weiner M, Conde MB, Jones B, et al. Treatment for preventing tuberculosis in children and adolescents: a randomized clinical trial of a 3-month, 12-dose regimen of a combination of rifapentine and isoniazid. JAMA Pediatr. 2015;169(3):247–55.

56. Moro R, Scott N, Vernon A, Goldberg S, Schwartzman K, Narita M, et al. Pregnancy Safety Assessment of 3 Months of Once-Weekly Rifapentine and Isoniazid and 9 Months of Daily Isoniazid: A Post-Hoc Analysis of the PREVENT TB and the iAdhere Trials. Am J Respir Crit Care Med. 2016;193:A7859. In: American Thoracic Society 2016 International Conference; 2016; San Francisco.

57. Belknap R, Holland D, Feng PJ, Millet JP, Cayla JA, Martinson NA, et al. Self-administered Versus Directly Observed Once-Weekly Isoniazid and Rifapentine Treatment of Latent Tuberculosis Infection: A Randomized Trial. Ann Intern Med. 2017;167(10):689–97.

58. Balcells ME, Thomas SL, Godfrey-Faussett P, Grant AD. Isoniazid preventive therapy and risk for resistant tuberculosis. Emerg Infect Dis. 2006;12(5):744–51.

59. den Boon S, Matteelli A, Getahun H. Rifampicin resistance after treatment for latent tuberculous infection: a systematic review and meta-analysis. Int J Tuberc Lung Dis. 2016;20(8):1065–71.

60. Podany AT, Bao Y, Swindells S, Chaisson RE, Andersen JW, Mwelase T, et al. Efavirenz Pharmacokinetics and Pharmacodynamics in HIV-Infected Persons Receiving Rifapentine and Isoniazid for Tuberculosis Prevention. Clin Infect Dis. 2015;61(8):1322–7.

61. Weiner M, Egelund EF, Engle M, Kiser M, Prihoda TJ, Gelfond JA, et al. Pharmacokinetic interaction of rifapentine and raltegravir in healthy volunteers. J Antimicrob Chemother. 2014;69(4):1079–85.

62. Sotgiu G, Matteelli A, Getahun H, Girardi E, Sañé Schepisi M, Centis R, et al. Monitoring toxicity in individuals receiving treatment for latent tuberculosis infection: a systematic review versus expert opinion. The European Respiratory Journal. 2015;45(4):1170–3.

63. WHO. Guidelines for treatment of drug-susceptible tuberculosis and patient care (2017 update). 2017.

64. Garcia-Prats AJ, Zimri K, Mramba Z, Schaaf HS, Hesseling AC. Children exposed to multidrug-resistant tuberculosis at a home-based day care centre: a contact investigation. Int J Tuberc Lung Dis. 2014;18(11):1292–8.

65. Trieu L, Proops DC, Ahuja SD. Moxifloxacin Prophylaxis against MDR TB, New York, New York, USA. Emerg Infect Dis. 2015;21(3):500–3.

66. Bamrah S, Brostrom R, Dorina F, Setik L, Song R, Kawamura LM, et al. Treatment for LTBI in contacts of MDR-TB patients, Federated States of Micronesia, 2009–2012. Int J Tuberc Lung Dis. 2014;18(8):912–8.

67. 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. Pediatrics. 2002;109(5):765–71.

68. WHO. Companion Handbook to the WHO Guidelines for the Programmatic Management of Drug-Resistant Tuberculosis. 2014.

69. WHO. Active tuberculosis drug-safety monitoring and management (aDSM): Framework for implementation. 2015.

9. REFERENCES


70. Takizawa T, Hashimoto K, Minami T, Yamashita S, Owen K. The comparative arthropathy of fluoroquinolones in dogs. Hum Exp Toxicol. 1999;18(6):392–9.

71. Hampel B, Hullmann R, Schmidt H. Ciprofloxacin in pediatrics: worldwide clinical experience based on compassionate use--safety report. Pediatr Infect Dis J. 1997;16(1):127–9; discussion 60–2.

72. Warren RW. Rheumatologic aspects of pediatric cystic fibrosis patients treated with fluoroquinolones. Pediatr Infect Dis J. 1997;16(1):118–22; discussion 23–6.

73. Denholm JT, Matteelli A, Reis A. Latent tuberculous infection: ethical considerations in formulating public health policy. Int J Tuberc Lung Dis. 2015;19(2):137–40.

74. WHO. Ethics guidance for the implementation of the End TB Strategy. 2017.

75. Kass NE. An ethics framework for public health. Am J Public Health. 2001;91(11):1776–82.

76. Getahun H, Matteelli A, Abubakar I, Hauer B, Pontali E, Migliori GB. Advancing global programmatic management of latent tuberculosis infection for at risk populations. European Respiratory Journal. 2016;47(5):1327–30.

39

Ann

ex 1.

GRA

DE

profi

le ta

bles

for n

ew re

com

men

datio

nsPI

CO 1:

Wha

t is

the

prev

alen

ce o

f LTB

I, ris

k of

pro

gres

sion

to a

ctiv

e TB

and

cum

ulat

ive

prev

alen

ce o

f act

ive

TB a

mon

g ho

useh

old

cont

acts

with

out H

IV in

diff

eren

t age

gro

ups

in h

igh

TB in

cide

nce

coun

trie

s?

Prev

alen

ce a

nd ri

sk o

f LTB

I in

HIV

-neg

ativ

e ho

useh

old

cont

acts

in h

igh-

TB in

cide

nce

coun

trie

sQ

ualit

y as

sess

men

tN

o. LT

BI+/

No.

test

edEff

ect

Qua

lity

Impo

rtan

ceN

o. o

f stu

dies

Des

ign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nC

ompa

rato

r0

–5 y

ears

RR(9

5% C

I)A

bsol

ute

per

100

0 (9

5% C

I)

AG

E G

ROU

PS C

OM

PARE

D: 5

–10

YEA

RS V

S 0

–5 Y

EARS

14 (1

–14)

C

ross

–se

ctio

nal

Not

ser

ious

1,2

Ser

ious

3N

ot s

erio

usN

ot s

erio

us4

2265

/85

0712

98/

9526

1.62

(1.2

5;2.

11)

85.1

(34.

2;15

1.1)

Mod

erat

eIm

port

ant

AG

E G

ROU

PS C

OM

PARE

D: 1

0–1

5 Y

EARS

VS

0–5

YEA

RS

11 (1

,3,5

,7-1

4)C

ross

-se

ctio

nal

Not

ser

ious

5 S

erio

us6

Not

ser

ious

Not

ser

ious

726

16/

6782

1093

/90

052.

33(1

.55;

3.50

)16

1.6(6

7.2;

303.

3)M

oder

ate

Impo

rtan

t

AG

E G

ROU

PS C

OM

PARE

D: 5

–15

YEA

RS V

S 0

–5 Y

EARS

16 (3

,5,8

, 10

,12,15

–25)

Cro

ss-

sect

iona

lSe

rious

8 S

erio

us9

Not

ser

ious

Not

ser

ious

1037

09/

8772

1605

/50

951.3

2(1

.11;1.

56)

99.7

(34.

9;17

6.5)

Low

Impo

rtan

t

AG

E G

ROU

PS C

OM

PARE

D: >

15 Y

EARS

VS

0–5

YEA

RS

19 (3

–5,8

–10,

12

–14,

16,17

,19

,20–

26)

Cro

ss-

sect

iona

lN

ot s

erio

us11

Ser

ious

12N

ot s

erio

usN

ot s

erio

us13

1321

8/21

962

1979

/67

632.

04(1

.53;

2.63

)29

3.9

(155

.1;47

5.7)

Mod

erat

eIm

port

ant

1 Po

tent

ial s

elec

tion

bias

in (2

), as

onl

y 69

% o

f par

ticip

ants

wer

e ho

useh

old

cont

acts

.2

Pote

ntia

l mis

clas

sific

atio

n: E

ight

stu

dies

(3–5

,7,10

,11,13

,14) d

id n

ot in

dica

te w

heth

er h

ouse

hold

con

tact

s w

ith a

ctiv

e TB

wer

e ex

clud

ed fr

om th

e an

alys

is o

r did

not

pro

vide

suffi

cien

t dat

a to

cal

cula

te th

e nu

mbe

r of h

ouse

hold

co

ntac

ts w

ith a

ctiv

e TB

per

age

str

atum

. 3

Hig

h he

tero

gene

ity a

mon

g st

udie

s (I

2 = 9

4%) p

roba

bly

due

to d

iffer

ence

in b

ackg

roun

d TB

inci

denc

e. R

isk

ratio

s of

two

stud

ies

(1,5

) sho

wed

opp

osite

effe

ct.

4 Sm

all s

ampl

e si

ze in

(5) (

n <

50).

5 Po

tent

ial m

iscl

assi

ficat

ion:

Sev

en s

tudi

es (3

,5,6

,10,11

,13,14

) di

d no

t ind

icat

e w

heth

er h

ouse

hold

con

tact

s w

ith a

ctiv

e TB

wer

e ex

clud

ed fr

om th

e an

alys

is o

r did

not

pro

vide

suffi

cien

t dat

a to

cal

cula

te th

e nu

mbe

r of h

ouse

hold

co

ntac

ts w

ith a

ctiv

e TB

per

age

str

atum

.6

Hig

h he

tero

gene

ity a

mon

g st

udie

s (I

2 = 9

7%) p

roba

bly

due

to th

e di

ffere

nces

in b

ackg

roun

d TB

inci

denc

e. R

isk

ratio

of o

ne s

tudy

(5) s

how

ed o

ppos

ite e

ffect

.7

Wid

e co

nfide

nce

inte

rval

of p

oole

d ris

k ra

tio. S

mal

l sam

ple

size

s in

(5) (

n <

50) a

nd (1

2) (n

< 10

0).

8 Po

tent

ial s

elec

tion

bias

in (1

5), a

s on

ly 8

9% o

f par

ticip

ants

wer

e ho

useh

old

cont

acts

.9

Hig

h he

tero

gene

ity a

mon

g st

udie

s (I

2 = 9

3%) p

roba

bly

due

to d

iffer

ence

s in

bac

kgro

und

TB in

cide

nce.

Ris

k ra

tios

in th

ree

stud

ies

show

ed o

ppos

ite e

ffect

s (5

,19,2

1).

10

Smal

l sam

ple

size

in (5

) and

(18)

(n <

50)

.11

Po

tent

ial m

iscl

assi

ficat

ion:

Ten

stu

dies

(3–

5,10

,13,14

,20,

21,2

3,26

) di

d no

t ind

icat

e w

heth

er h

ouse

hold

con

tact

s w

ith a

ctiv

e TB

wer

e ex

clud

ed fr

om th

e an

alys

is o

r did

not

pro

vide

suffi

cien

t dat

a to

cal

cula

te th

e nu

mbe

r of

hous

ehol

d co

ntac

ts w

ith a

ctiv

e TB

per

age

str

atum

.12

H

igh

hete

roge

neity

am

ong

stud

ies

(I2 =

98%

) pro

babl

y du

e to

diff

eren

ces

in b

ackg

roun

d TB

inci

denc

e.

13

Smal

l sam

ple

size

s in

(5) a

nd (2

6) (n

< 10

0).

ANNEX 1. GRADE PROFILE TABLES FOR NEW RECOMMENDATIONS


Dev

elop

men

t of a

ctiv

e TB

in h

ouse

hold

con

tact

s w

ith LT

BI in

hig

h TB

inci

denc

e co

untr

ies

Qua

lity

asse

ssm

ent

No

of c

onta

cts

(act

ive

TB/n

o. LT

BI)

Effec

t

Qua

lity

Impo

rtan

ceN

o, o

f stu

dies

Des

ign

Risk

of b

ias

Lim

itatio

nsIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Com

para

tor

0–5

yea

rsRR

(9

5% C

I)

Abs

olut

e pe

r 10

00

(9

5% C

I)

AG

E G

ROU

PS C

OM

PARE

D: 5

–15

YEA

RS V

S 0

–5 Y

EARS

4 (8

,13,15

,16)

Coh

ort

Not

ser

ious

Not

ser

ious

Serio

us1

Not

ser

ious

Serio

us2

54/1

329

73/6

300.

28(0

.12;0

.65)

83.8

(4

0.3;

102.

3)Lo

wC

ritic

al

AG

E G

ROU

PS C

OM

PARE

D: >

15 Y

EARS

VS

0–5

YEA

RS

3 (8

,13,16

)C

ohor

tN

ot s

erio

usN

ot s

erio

usSe

rious

3N

ot s

erio

usN

ot s

erio

us18

6/47

4673

/595

0.22

(0

.08;

0.60

)95

.5 (4

9.1;1

12.6

)M

oder

ate

Crit

ical

Beca

use

of th

e sm

all n

umbe

r of s

tudi

es in

the

othe

r cat

egor

ies,

onl

y da

ta fr

om s

tudi

es w

ith a

follo

w-u

p of

1–2

year

s in

hig

h TB

inci

denc

e co

untr

ies

are

pres

ente

d in

the

tabl

e.

1 Se

rious

inco

nsis

tenc

ies

due

to h

eter

ogen

eity

(I2 =

71%

): O

ne s

tudy

sho

wed

an

incr

ease

d ris

k in

the

age

grou

p 5–

15 y

ears

. Thi

s w

as n

ot o

bser

ved

in th

e ot

her s

tudi

es.

2 Sm

all n

umbe

r of e

vent

s.3

H

igh

hete

roge

neity

am

ong

stud

ies

prob

ably

due

to d

iffer

ence

s in

bac

kgro

und

TB in

cide

nce

and

met

hods

use

d to

dia

gnos

e ac

tive

TB (I

2 = 8

9.3%

).

Cum

ulat

ive

prev

alen

ce o

f act

ive

TB in

hou

seho

ld c

onta

cts

irres

pect

ive

of b

asel

ine

LTBI

sta

tus

in h

igh

TB in

cide

nce

coun

trie

s

Qua

lity

asse

ssm

ent

No

of c

onta

cts

(act

ive

TB/t

otal

no.

con

tact

s)Eff

ect

Qua

lity

Impo

rtan

ceN

o. o

f stu

dies

Des

ign

Risk

of b

ias

Lim

itatio

nsIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Com

para

tor

0–5

yea

rsRR

(9

5% C

I)

Abs

olut

e pe

r 10

00

(95%

CI)

AG

E G

ROU

PS C

OM

PARE

D: 5

–15

YEA

RS V

S 0

–5 Y

EARS

6 (8

,13,15

,16

,18,2

7)1

Coh

ort

Not

ser

ious

Not

se

rious

Serio

us2

Not

ser

ious

Not

ser

ious

131/

4389

203/

2903

0.39

(0

.18;0

.85)

42.9

(1

0.6;

57.6

)M

oder

ate

Impo

rtan

t

AG

E G

ROU

PS C

OM

PARE

D: >

15 Y

EARS

VS

0–5

YEA

RS

4 (8

,13,16

,27)

Coh

ort

Not

ser

ious

Not

se

rious

Not

ser

ious

Not

ser

ious

Not

ser

ious

417/

1085

619

2/27

640.

68

(0.5

6;0.

83)

22

(12.

1;30.

3)H

igh

Impo

rtan

t

Ow

ing

to th

e sm

all n

umbe

r of s

tudi

es in

the

othe

r cat

egor

ies,

onl

y da

ta fr

om s

tudi

es w

ith a

follo

w-u

p of

1–2

year

s in

hig

h TB

inci

denc

e co

untr

ies

are

pres

ente

d in

the

tabl

e.

1 O

ne o

utlie

r (28

) was

exc

lude

d be

caus

e of

unc

erta

inty

abo

ut th

e ca

ses

incl

uded

(co-

prev

alen

t vs

inci

dent

cas

es).

2

Hig

h he

tero

gene

ity a

mon

g st

udie

s (I

2 = 8

7.6%

), pr

obab

ly d

ue to

the

diffe

renc

e in

bac

kgro

und

TB in

cide

nce.

41

Act

ive

TB in

hou

seho

ld c

onta

cts

with

LTBI

and

in th

e ge

nera

l pop

ulat

ion

in h

igh-

TB in

cide

nce

coun

trie

s (1

2 m

onth

s)

ACT

IVE

TB D

ISEA

SE IN

HO

USE

HO

LD C

ON

TACT

S W

ITH

LTBI

INFE

CTIO

N IN

HIG

H T

B IN

CID

ENCE

CO

UN

TRIE

SCO

MPA

RISO

N W

ITH

TH

E G

ENER

AL

POPU

LATI

ON

(FO

LLO

W-U

P O

F 12

MO

NTH

S)

Qua

lity

asse

ssm

ent

No.

of c

onta

cts

(act

ive

TB/n

o. LT

BI)

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f stu

dies

Des

ign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nC

ompa

rato

rG

ener

al

popu

latio

n1 RR

(9

5% C

I)

Abs

olut

e

per 1

00

0 (9

5% C

I)

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

0–5

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

2 (8

,15)

Coh

ort

Ser

ious

2Se

rious

3N

ot s

erio

usVe

ry s

erio

us4

0/35

41/1

0 00

024

.32

(0.7

3;81

1.02)

63

(–0.

7;21

87.1)

Ver

y lo

wC

ritic

al32

/230

13/1

0 00

0

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–9

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Ser

ious

2N

ot s

erio

usN

ot s

erio

usSe

rious

612

/298

13/1

0 00

030

.98

(14.

26;6

7.31

)39

(17.

2;86

.2)

Low

Crit

ical

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

10–1

4 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

1 (8)

Coh

ort

Ser

ious

2N

ot s

erio

usN

ot s

erio

usSe

rious

626

/363

13

/10

000

55.1

(28.

55;10

6.33

)70

.3

(35.

8;13

6.9)

Low

Crit

ical

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–1

5 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

2 (8

,15)

Coh

ort

Ser

ious

2N

ot s

erio

us5

Not

ser

ious

Serio

us6

4/67

41/1

0 00

027

.13

(17.

47;5

4.07

)70

.5

(21.3

;220

.7)

Low

Crit

ical

38/6

6113

/10

000

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

> 15

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Ser

ious

3N

ot s

erio

usN

ot s

erio

usSe

rious

615

5/38

7913

/10

000

30.7

4 (1

7.46

;54.

07)

38.7

(21.4

;69)

Low

Crit

ical

1 LT

BI d

oes

not a

pply

to th

e ge

nera

l pop

ulat

ion.

2 A

scer

tain

men

t bia

s hi

ghly

like

ly, a

s TB

cas

es in

the

gene

ral p

opul

atio

n de

tect

ed p

assi

vely

, whi

le T

B ca

ses

in c

onta

cts

dete

cted

act

ivel

y. A

s a

resu

lt, th

e re

lativ

e an

d ab

solu

te ri

sks

mig

ht b

e ov

eres

timat

ed. T

he c

ompo

sitio

n of

th

e ge

nera

l and

the

stud

y po

pula

tion

diffe

red

(gen

eral

pop

ulat

ion

of a

ll ag

es v

ersu

s a

spec

ific

age

grou

p).

3

Hig

h he

tero

gene

ity a

mon

g st

udie

s (I

2 = 8

3.9%

), pr

obab

ly d

ue to

diff

eren

ces

in b

ackg

roun

d TB

inci

denc

e.4

Se

rious

impr

ecis

ion

with

a w

ide

confi

denc

e in

terv

al fo

r the

effe

ct e

stim

ates

, pro

babl

y du

e to

sm

all s

tudy

siz

e an

d nu

mbe

r of o

utco

me

even

ts.

5

I2 = 7

2.5%

, ind

icat

ing

mod

erat

e he

tero

gene

ity, p

roba

bly

due

to d

iffer

ence

s in

bac

kgro

und

TB p

reva

lenc

e; h

owev

er, t

here

is a

tren

d ac

ross

age

gro

ups

and

stud

ies.

6

Fe

w e

vent

s an

d w

ide

CI.



Act

ive

TB in

hou

seho

ld c

onta

cts

with

LTBI

com

pare

d w

ith g

ener

al p

opul

atio

n in

hig

h-TB

inci

denc

e co

untr

ies

(24

mon

ths)

ACT

IVE

TB D

ISEA

SE IN

HO

USE

HO

LDS

OF

CON

TACT

S W

ITH

LTBI

INFE

CTIO

N IN

HIG

H-T

B IN

CID

ENCE

CO

UN

TRIE

SCO

MPA

RISO

N W

ITH

TH

E G

ENER

AL

POPU

LATI

ON

(FO

LLO

W-U

P ≤

24 M

ON

THS)

1

Qua

lity

asse

ssm

ent

No

of c

onta

cts

(act

ive

TB/n

o. LT

BI)

Effec

tQ

ualit

yIm

port

ance

No.

of s

tudi

esD

esig

nRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Com

para

tor

Gen

eral

pop

2RR

(9

5% C

I)A

bsol

ute

per

100

0 (9

5% C

I)

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

0–5

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

3 (8

,15,16

)C

ohor

t S

erio

us3

Ser

ious

4N

ot s

erio

usSe

rious

5

0/35

82/1

0 00

022

.87

(7.6

5;68

.63)

108.

6 (3

3;33

4.6)

Ver

y lo

wIm

port

ant

26/3

2041

/10

000

32/2

3026

/10

000

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–9

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Ser

ious

3N

ot s

erio

usN

ot s

erio

usSe

rious

512

/298

26/1

0 00

015

.49

(7.8

9;30

.4)

37.7

(1

7.9;

76.4

)Lo

wIm

port

ant

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

10–1

4 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

1 (8)

Coh

ort

Ser

ious

3N

ot s

erio

usN

ot s

erio

usSe

rious

526

/363

26/1

0 00

027

.55

(1

6.16

;46.

96)

69

(39.

4;11

9.5)

Low

Impo

rtan

t

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–1

5 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

3 (8

,15,16

)C

ohor

t S

erio

us3

Serio

us6

Not

ser

ious

Serio

us5

4/67

82/1

0 00

08.

22

(2.3

;29.

36)

35.8

(6

.5;14

0.8)

Ver

y lo

wIm

port

ant

6/47

541

/10

000

38/6

6126

/10

000

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

OV

ER 15

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

2 (8

,16)

Coh

ort

Ser

ious

3N

ot s

erio

us7

Not

ser

ious

Not

ser

ious

26/5

7141

/10

000

13.3

5

(9.4

6;18

.83)

41.4

(2

8.3;

59.7

)M

oder

ate

Impo

rtan

t15

5/38

7926

/10

000

1 Th

ese

com

paris

ons

incl

uded

stu

dies

with

a m

axim

um fo

llow

-up

of 2

4 m

onth

s; th

eref

ore,

TB

inci

denc

e in

the

gene

ral p

opul

atio

n w

as m

ultip

lied

by a

fact

or o

f 2 to

est

imat

e th

e nu

mbe

r of c

ases

occ

urrin

g du

ring

24 m

onth

s.

2 LT

BI d

oes

not a

pply

to th

e ge

nera

l pop

ulat

ion.

3 A

scer

tain

men

t bia

s hi

ghly

like

ly: T

B ca

ses

in th

e ge

nera

l pop

ulat

ion

dete

cted

pas

sive

ly, w

hile

TB

case

s in

the

cont

acts

det

ecte

d ac

tivel

y. A

s a

resu

lt, re

lativ

e an

d ab

solu

te ri

sks

mig

ht b

e ov

eres

timat

ed. T

he c

ompo

sitio

n of

the

gene

ral a

nd s

tudy

pop

ulat

ions

diff

ers

(gen

eral

pop

ulat

ion

of a

ll ag

es v

ersu

s a

spec

ific

age

grou

p). T

B in

cide

nce

in th

e po

pula

tion

was

est

imat

ed b

y m

ultip

lyin

g th

e ye

arly

not

ifica

tion

rate

by

a fa

ctor

of 2

. 4

Hig

h he

tero

gene

ity b

etw

een

stud

ies

prob

ably

due

to d

iffer

ence

in b

ackg

roun

d TB

inci

denc

e (I

2 = 8

4.4%

). 5

Few

eve

nts

and

wid

e C

I. 6

I2 = 8

8.1%

, ind

icat

ing

high

het

erog

enei

ty p

roba

bly

due

to d

iffer

ence

in b

ackg

roun

d TB

pre

vale

nce;

how

ever

, the

re is

a tr

end

acro

ss a

ge g

roup

s an

d st

udie

s.

7 I2 =

16%

.

43

Act

ive

TB in

hou

seho

ld c

onta

cts

irres

pect

ive

of LT

BI s

tatu

s co

mpa

red

with

gen

eral

pop

ulat

ion

in h

igh

TB in

cide

nce

coun

trie

s

(12

mon

ths)

CUM

ULA

TIV

E PR

EVA

LEN

CE O

F A

CTIV

E TB

IN H

OU

SEH

OLD

CO

NTA

CTS

IRRE

SPEC

TIV

E O

F BA

SELI

NE

LTBI

STA

TUS

IN H

IGH

TB

INCI

DEN

CE C

OU

NTR

IES

COM

PARI

SON

WIT

H T

HE

GEN

ERA

L PO

PULA

TIO

N (F

OLL

OW

-UP

OF

12 M

ON

THS)

Qua

lity

asse

ssm

ent

No.

of c

onta

cts

(act

ive

TB/t

otal

)Eff

ect

Qua

lity

Impo

rtan

ceN

o of

stu

dies

Des

ign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nC

ompa

rato

rG

ener

al p

opRR

(9

5% C

I)

Abs

olut

e ris

k pe

r 10

00

(95%

CI)

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

0–5

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

3 (8

,15,18

)C

ohor

tSe

rious

1N

ot s

erio

us2

Not

ser

ious

Serio

us3

2/31

28/1

0 00

025

.86

(16.

87;3

9.66

)68

(4

3.4;

105.

7)Lo

wIm

port

ant

9/10

841

/10

000

73/1

791

13/1

0 00

0

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–9

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Serio

us1

Not

ser

ious

Not

ser

ious

Serio

us3

35/1

464

13/1

0 00

018

.39

(9.7

5;34

.68)

22.6

(1

1.4;4

3.8)

Low

Impo

rtan

t

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

10–1

4 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

1 (8)

Coh

ort

Serio

us1

Not

ser

ious

Not

ser

ious

Serio

us3

45/1

340

13/1

0 00

025

.83

(13.

97;4

7.76

)32

.3

(16.

9;60

.8)

Low

Impo

rtan

t

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–1

5 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

3 (8

,15,18

)C

ohor

tSe

rious

1N

ot s

erio

us2

Not

ser

ious

Serio

us3

8/10

228

/10

000

24.11

(1

6.89

;34.

43)

63.2

(4

3.4;

91.4

)Lo

wIm

port

ant

16/1

6141

/10

000

80/2

804

13/1

0 00

0

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

OV

ER 15

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Serio

us1

Not

ser

ious

Not

ser

ious

Not

ser

ious

301/

9380

13/1

0 00

024

.68

(14.

18;4

2.98

)30

.8

(17.

1;54.

6)M

oder

ate

Impo

rtan

t

1 A

scer

tain

men

t bia

s hi

ghly

like

ly, a

s TB

cas

es in

the

gene

ral p

opul

atio

n de

tect

ed p

assi

vely

, whi

le T

B ca

ses

in th

e co

ntac

ts d

etec

ted

activ

ely.

As

a re

sult,

the

rela

tive

and

abso

lute

risk

mig

ht b

e ov

eres

timat

ed. T

he c

ompo

sitio

n of

the

gene

ral a

nd s

tudy

pop

ulat

ions

diff

ers

(gen

eral

pop

ulat

ion

of a

ll ag

es v

ersu

s a

spec

ific

age

grou

p).

2 I2 =

0%

. 3

Few

eve

nts

and

wid

e C

I.



Activ

e TB

in h

ouse

hold

con

tact

s irr

espe

ctive

of L

TBI s

tatu

s co

mpa

red

with

gen

eral

pop

ulat

ion

in h

igh-

TB in

ciden

ce c

ount

ries

(24

mon

ths)

CUM

ULA

TIV

E PR

EVA

LEN

CE O

F A

CTIV

E TB

IN H

OU

SEH

OLD

CO

NTA

CTS

IRRE

SPEC

TIV

E O

F BA

SELI

NE

LTBI

STA

TUS

IN H

IGH

-TB

INCI

DEN

CE C

OU

NTR

IES

COM

PARI

SON

WIT

H T

HE

GEN

ERA

L PO

PULA

TIO

N (F

OLL

OW

-UP

OF

24 M

ON

THS)

1

Qua

lity

asse

ssm

ent

No

of c

onta

cts

(act

ive

TB/t

otal

no.

con

tact

s)Eff

ect

Qua

lity

Impo

rtan

ceN

o. o

f stu

dies

Des

ign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nC

ompa

rato

rG

ener

al

popu

latio

nRR

(9

5% C

I)A

bsol

ute

risk

per

100

0 (9

5% C

I)

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

0–5

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

5 (8

,15,16

,18

,27)

Coh

ort

Serio

us2

Not

ser

ious

3N

ot s

erio

usSe

rious

4

2/31

55/1

0 00

0

14.8

(9

.82;

22.3

)83

.9

(53.

6;12

9.5)

Low

Impo

rtan

t

37/3

3510

0/10

000

9/10

882

/10

000

55/5

0841

/10

000

73/1

791

26/1

0 00

0

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–9

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

1 (8)

Coh

ort

Serio

us2

Not

ser

ious

Not

ser

ious

Serio

us4

35/1

464

26/1

0 00

09.

2 (5

.55;

15.2

3)21

.3

(11.8

;37)

Low

Impo

rtan

t

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

10–1

4 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

1 (8)

Coh

ort

Serio

us2

Not

ser

ious

Not

ser

ious

Serio

us4

45/1

340

26/1

0 00

012

.92

(8.0

;20.

86)

31

(18.

2;51

.6)

Low

Impo

rtan

t

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

5–1

5 Y

EARS

VS

GEN

ERA

L PO

PULA

TIO

N

5 (8

,15,16

,18

,27)

Coh

ort

Serio

us2

Serio

us5

Not

ser

ious

Not

ser

ious

8/10

255

/10

000

6.29

(2

.88;

13.7

2)32

.2

(11.4

;77.

4)Lo

w

Impo

rtan

t

5/43

910

0/10

000

16/1

6182

/10

000

10/6

9141

/10

000

80/2

804

26/1

0 00

0

COM

PARI

SON

: HO

USE

HO

LD C

ON

TACT

S A

GED

OV

ER 15

YEA

RS V

S G

ENER

AL

POPU

LATI

ON

3 (8

,16,2

7)C

ohor

tSe

rious

2N

ot s

erio

us6

Not

ser

ious

Not

ser

ious

34/4

3210

0/10

000

11.6

7 (7

.55;

18.0

2)59

.4

(36.

5;94

.7)

Mod

erat

eIm

port

ant

49/7

1941

/100

00

301/

9380

26/1

0000

1 Th

ese

com

paris

ons a

re b

ased

on

stud

ies w

ith a

max

imum

follo

w-u

p of

24

mon

ths;

ther

efor

e, T

B in

cide

nce

in th

e ge

nera

l pop

ulat

ion

was

mul

tiplie

d by

a fa

ctor

of 2

to e

stim

ate

the

num

ber o

f cas

es o

ccur

ring

durin

g 24

mon

ths.

2

Asc

erta

inm

ent b

ias

high

ly li

kely

, as

TB c

ases

in th

e ge

nera

l pop

ulat

ion

dete

cted

pas

sive

ly, w

hile

TB

case

s in

the

cont

acts

det

ecte

d ac

tivel

y. A

s a

resu

lt, th

e re

lativ

e an

d ab

solu

te ri

sks

mig

ht b

e ov

eres

timat

ed. T

he c

ompo

sitio

n of

the

gene

ral a

nd s

tudy

pop

ulat

ions

diff

ers

(gen

eral

pop

ulat

ion

of a

ll ag

es v

ersu

s a

spec

ific

age

grou

p). T

B in

cide

nce

in th

e po

pula

tion

was

est

imat

ed b

y m

ultip

lyin

g th

e ye

arly

not

ifica

tion

rate

by

a fa

ctor

of 2

. 3

M

oder

ate

hete

roge

neity

am

ong

stud

ies

(I2 =

67.

1%) p

roba

bly

due

to d

iffer

ence

s in

bac

kgro

und

TB in

cide

nce.

4

Few

eve

nts

and

wid

e C

I. 5

H

igh

hete

roge

neity

am

ong

stud

ies

(I2 =

87.

5%) p

roba

bly

due

to d

iffer

ence

s in

bac

kgro

und

TB in

cide

nce.

6

M

oder

ate

hete

roge

neity

am

ong

stud

ies

(I2 =

72.

5%) p

roba

bly

due

to d

iffer

ence

s in

bac

kgro

und

TB in

cide

nce.

45

PICO

2: W

hat i

s th

e ac

cura

cy o

f WH

O s

ympt

omat

ic s

cree

ning

to e

xclu

de a

ctiv

e TB

in in

divi

dual

s w

ith H

IV o

n an

tiret

rovi

ral

trea

tmen

t (A

RT)?

Four

-sym

ptom

scr

eeni

ng p

lus

ches

t rad

iogr

aphi

c fin

ding

s to

exc

lude

act

ive

TB in

indi

vidu

als

with

HIV

Popu

latio

n: A

dults

and

ado

lesc

ents

with

HIV

on

ART

Sens

itivi

ty

0.85

(95%

CI:

0.70

;0.9

3)

Spec

ifici

ty

0.30

(95%

CI:

0.26

;0.3

3)Pr

eval

ence

1%

5%10

%

Out

com

eN

os o

f stu

dies

an

d pa

tient

sSt

udy

desi

gn

Fact

ors

that

may

dec

reas

e qu

ality

of e

vide

nce

Effec

t per

100

0 pa

tient

s te

sted

Test

ac

cura

cy

Qua

lity

of

evid

ence

Risk

of b

ias

Indi

rect

ness

Inco

nsis

tenc

yIm

prec

isio

nPu

blic

atio

n bi

as

Pre-

test

pr

obab

ility

, 1%

Pre-

test

pr

obab

ility

, 5%

Pre-

test

pr

obab

ility

, 10

%

True

pos

itive

s (p

atie

nts

with

act

ive

TB)

2 st

udie

s 64

6 pa

tient

s

Cro

ss-s

ectio

nal

(coh

ort t

ype

accu

racy

stu

dy)

Not

ser

ious

N

ot s

erio

us

Not

ser

ious

Se

rious

1N

one2

8 (7

–9)

42 (3

5–46

)85

(70

–93)

㊉㊉

㊉◯

M

oder

ate

Fals

e ne

gativ

es

(pat

ient

s in

corr

ectly

cl

assi

fied

as n

ot h

avin

g ac

tive

TB)

2 (1

–3)

8 (4

–15)

15 (

7–30

)

True

neg

ativ

es (p

atie

nts

with

out a

ctiv

e TB

) 2

stud

ies

646

patie

nts

Cro

ss-s

ectio

nal

(coh

ort t

ype

accu

racy

stu

dy)

Not

ser

ious

N

ot s

erio

us

Not

ser

ious

N

ot s

erio

us

Non

e2

295

(260

–327

)28

3 (2

50–3

14)

268

(237

–297

)㊉

㊉㊉

㊉

Hig

h Fa

lse

posi

tives

(pat

ient

s in

corr

ectly

cla

ssifi

ed a

s ha

ving

act

ive

TB)

695

(663

–730

)66

7 (6

36–7

00)

632

(603

–663

)

From

refe

renc

es (2

9,30

)1

Impr

ecis

e es

timat

e fo

r sen

sitiv

ity. D

owng

rade

d by

one

.2

Th

e po

ssib

ility

of p

ublic

atio

n bi

as is

not

exc

lude

d, b

ut it

was

not

con

side

red

of s

uffici

ent c

once

rn to

dow

ngra

de.



PICO

3: W

hat i

s th

e ac

cura

cy o

f sym

ptom

atic

scr

eeni

ng a

nd/o

r che

st x

-ray

to e

xclu

de a

ctiv

e TB

in c

onta

cts

of p

ulm

onar

y TB

cas

es

with

out H

IV in

hig

h TB

inci

denc

e co

untr

ies?

Ch

est r

adio

grap

hic

findi

ngs

for e

xclu

sion

of a

ctiv

e TB

in c

onta

cts

of T

B ca

ses

with

out H

IV in

hig

h-TB

inci

denc

e co

untr

ies

Inde

x te

st: C

hest

X-r

ay. A

ny a

bnor

mal

ity |

Refe

renc

e te

st: S

putu

m c

ultu

re a

nd/o

r sm

ear

Plac

e of

test

ing:

Tria

ge

Test

–tre

atm

ent p

athw

ay: C

hest

X-r

ay p

ositi

ve ➞

con

firm

ator

y te

st (m

ycob

acte

rial c

ultu

re o

r Gen

eXpe

rt) ➞

ant

i-TB

chem

othe

rapy

(6–9

mon

ths’

ant

ibio

tics)

Out

com

eN

os o

f st

udie

s an

d pa

tient

s St

udy

desi

gnFa

ctor

s th

at m

ay d

ecre

ase

qual

ity o

f evi

denc

eEff

ect p

er 10

0 0

00

Sens

itivi

ty: 0

.94

(95%

CI:

0.86

;0.9

8)Sp

ecifi

city

: 0.8

7 (9

5% C

I: 0.

80;0

.92)

Qua

lity

of

evid

ence

Risk

of b

ias

Indi

rect

ness

Inco

nsis

tenc

yIm

prec

isio

nPu

blic

atio

n bi

as

True

pos

itive

s

(pat

ient

s w

ith a

ctiv

e TB

) 7

stud

ies

251 4

10

patie

nts

Cro

ss-

sect

iona

l (c

ohor

t typ

e ac

cura

cy

stud

y)

Serio

us1

Not

ser

ious

2N

ot s

erio

us3

Not

ser

ious

4N

one5

Prev

alen

ce (2

%):

1882

(171

6;19

54)

Prev

alen

ce (5

%):

4705

(429

0;48

85㊉

㊉㊉

◯

Mod

erat

eFa

lse

nega

tives

(pat

ient

s in

corr

ectly

cla

ssifi

ed a

s no

t hav

ing

activ

e TB

)

Prev

alen

ce (2

%) :

118

(46;

284)

Prev

alen

ce (5

%):

295

(115

;710

)

True

neg

ativ

es (p

atie

nts

with

out a

ctiv

e TB

) 7

stud

ies

251 4

10

patie

nts

Cro

ss-

sect

iona

l (c

ohor

t typ

e ac

cura

cy

stud

y)

Serio

us1

Not

ser

ious

2N

ot s

erio

us3

Not

ser

ious

4N

one5

Prev

alen

ce (2

%) :

85

064

(78

106;

89 8

66)

Prev

alen

ce (5

%):

82 4

60 (

75

715;

87 11

5)㊉

㊉㊉

◯

Mod

erat

e Fa

lse

posi

tives

(pat

ient

s in

corr

ectly

cla

ssifi

ed a

s ha

ving

act

ive

TB)

Prev

alen

ce (2

%) :

12 9

36 (8

134;

19

894)

Prev

alen

ce (5

%):

12 5

40 (

7 88

5;19

28

5)

From

refe

renc

es (3

1–37

)1

Lim

itatio

ns in

stu

dy d

esig

n (s

ee Q

UA

DA

S-2)

: Hig

h ris

k of

sel

ectio

n bi

as in

one

stu

dy (3

1). I

n al

l stu

dies

, les

s th

an h

alf o

f par

ticip

ants

rece

ived

the

refe

renc

e st

anda

rd; a

ccur

acy

was

cal

cula

ted

unde

r the

ass

umpt

ion

that

thos

e w

ho d

id n

ot re

ceiv

e th

e re

fere

nce

stan

dard

wer

e cu

lture

and

/or s

mea

r neg

ativ

e (n

o ac

tive

TB).

2

Indi

rect

ness

(see

QU

AD

AS-

2): S

ome

conc

ern

abou

t app

licab

ility

of r

efer

ence

sta

ndar

d in

2 s

tudi

es –

no

dow

ngra

ding

. 3

In

cons

iste

ncy:

Litt

le h

eter

ogen

eity

for s

ensi

tivity

and

spe

cific

ity (b

ased

on

visu

al in

spec

tion

of C

Is).

4

Impr

ecis

ion:

Pre

cise

est

imat

es fo

r sen

sitiv

ity a

nd s

peci

ficity

. 5

Pu

blic

atio

n bi

as: N

ot a

pplic

able

(the

evi

denc

e ba

se fo

r pub

licat

ion

bias

in s

tudi

es o

f dia

gnos

tic te

st a

ccur

acy

is v

ery

limite

d).

47

Any

sym

ptom

for e

xclu

sion

of a

ctiv

e TB

in c

onta

cts

of T

B ca

ses

with

out H

IV in

hig

h-TB

inci

denc

e co

untr

ies.

In

dex

text

: Any

sym

ptom

| Re

fere

nce

test

: Spu

tum

cul

ture

and

/or s

mea

r Pl

ace

of te

stin

g: T

riage

Te

st–t

reat

men

t pat

hway

: Sym

ptom

pos

itive

➞ c

onfir

mat

ory

test

(myc

obac

teria

l cul

ture

or G

eneX

pert

) ➞ a

nti-T

B ch

emot

hera

py (6

–9 m

onth

s’ a

ntib

iotic

s)

Out

com

eN

os o

f stu

dies

an

d pa

tient

sSt

udy

desi

gn

Fact

ors

that

may

dec

reas

e qu

ality

of e

vide

nce

Effec

t per

100

00

0 Se

nsiti

vity

: 0.7

3 (9

5% C

I: 0.

64;0

.80)

Sp

ecifi

city

: 0.7

7 (9

5% C

I: 0.

61;0

.87)

Qua

lity

of

evid

ence

Risk

of b

ias

Indi

rect

ness

Inco

nsis

tenc

yIm

prec

isio

nPu

blic

atio

n bi

as

True

pos

itive

s (p

atie

nts

with

act

ive

TB)

11 s

tudi

es

357

609

patie

nts

Cro

ss-s

ectio

nal

(coh

ort t

ype

accu

racy

stu

dy)

Very

ser

ious

1N

ot s

erio

us2

Not

ser

ious

3N

ot s

erio

us4

Non

e5

Prev

alen

ce (2

%):

1460

(1

282;

1608

) Pr

eval

ence

(5%

): 36

50

(320

5;40

20)

㊉㊉

◯◯

Lo

w

Fals

e ne

gativ

es

(pat

ient

s in

corr

ectly

cl

assi

fied

as n

ot

havi

ng a

ctiv

e TB

)

Prev

alen

ce (2

%):

540

(392

;718

)

Prev

alen

ce (5

%):1

350

(980

;1795

)

True

neg

ativ

es

(pat

ient

s w

ithou

t ac

tive

TB)

11 s

tudi

es

357

609

patie

nts

Cro

ss-s

ectio

nal

(coh

ort t

ype

accu

racy

stu

dy)

Very

ser

ious

1N

ot s

erio

us2

Serio

us3

Serio

us4

Non

e5

Prev

alen

ce (2

%):7

4 97

0 (6

0 07

4;85

260

)Pr

eval

ence

(5%

):72

675

(58

235;

82 6

50)

㊉◯

◯◯

Ve

ry lo

w

Fals

e po

sitiv

es

(pat

ient

s in

corr

ectly

cl

assi

fied

as h

avin

g ac

tive

TB)

Prev

alen

ce (2

%):2

3 03

0 (1

2 74

0;37

926

) Pr

eval

ence

(5%

):22

325

(12

350;

36 7

65)

From

refe

renc

es (3

1–34

,36,

38–4

3)1

Lim

itatio

ns in

stu

dy d

esig

n (s

ee Q

UA

DA

S-2)

: hig

h ris

k of

sel

ectio

n bi

as in

1 st

udy

(den

Boo

n, 2

006

) and

in tw

o st

udie

s unc

lear

risk

of b

ias f

or th

e re

fere

nce

stan

dard

. In

9 of

the

11 s

tudi

es le

ss th

an h

alf t

he p

artic

ipan

ts re

ceiv

ed

the

refe

renc

e st

anda

rd; a

ccur

acy

was

cal

cula

ted

unde

r the

ass

umpt

ion

that

thos

e w

ho d

id n

ot re

ceiv

e th

e re

fere

nce

stan

dard

wer

e cu

lture

and

/or s

mea

r neg

ativ

e (n

o ac

tive

TB) .

2

Indi

rect

ness

(see

QU

AD

AS-

2): N

o m

ajor

con

cern

abo

ut a

pplic

abili

ty.

3

Inco

nsis

tenc

y: M

oder

ate

hete

roge

neity

for s

ensi

tivity

and

sig

nific

ant h

eter

ogen

eity

for s

peci

ficity

(bas

ed o

n vi

sual

insp

ectio

n of

CIs

) – d

owng

radi

ng o

n sp

ecifi

city

. 4

Im

prec

isio

n: P

reci

se e

stim

ates

for s

ensi

tivity

and

impr

ecis

e es

timat

e fo

r spe

cific

ity.

5

Publ

icat

ion

bias

: Not

app

licab

le (t

he e

vide

nce

base

for a

sses

sing

pub

licat

ion

bias

in s

tudi

es o

f dia

gnos

tic te

st a

ccur

acy

is v

ery

limite

d.



PICO

4: C

ould

inte

rfer

on-g

amm

a re

leas

e as

says

be

used

as

an a

ltern

ativ

e to

tube

rcul

in s

kin

test

s to

iden

tify

indi

vidu

als

mos

t at

risk

of p

rogr

essi

on fr

om LT

BI to

act

ive

TB in

hig

h TB

inci

denc

e se

ttin

gs?

TST

or IG

RA fo

r ide

ntify

ing

indi

vidu

als

at g

reat

est r

isk

of p

rogr

essi

on to

act

ive

TB

Hea

d-to

hea

d-ev

alua

tions

of T

ST a

nd IG

RA (N

= 5

) Re

view

que

stio

n: A

mon

g pe

ople

at h

igh

risk

of LT

BI w

ho a

re n

ot tr

eate

d w

ith T

B pr

even

tive

ther

apy,

whi

ch te

st (e

.g. T

ST o

r IG

RA),

whe

n po

sitiv

e, c

an b

est i

dent

ify in

divi

dual

s m

ost a

t ris

k of

pr

ogre

ssio

n?

Out

com

e: P

redi

ctiv

e ut

ility

of t

he T

ST v

s co

mm

erci

al IG

RAs

for p

rogr

essi

on to

act

ive

TB

Patie

nts/

popu

latio

n: L

ongi

tudi

nal s

tudi

es o

f adu

lts a

nd c

hild

ren

with

out a

ctiv

e TB

at b

asel

ine

not t

reat

ed w

ith p

reve

ntiv

e th

erap

ySe

ttin

g: C

omm

unity

coh

orts

, ind

ivid

uals

att

endi

ng o

utpa

tient

clin

ics

(e.g

. peo

ple

livin

g w

ith H

IV),

indi

vidu

als

part

icip

atin

g in

RC

Ts, h

ouse

hold

con

tact

s; a

ll in

hig

h-in

cide

nce

coun

trie

s In

dex

test

: TST

(RT2

3 pu

rified

pro

tein

der

ivat

ive

or p

urifi

ed p

rote

in d

eriv

ativ

e S)

and

/or c

omm

erci

al b

lood

-bas

ed IG

RAs

(Qua

ntiF

ERO

N®-

TB G

old

In-T

ube

and

T-SP

OT®

.TB)

Im

port

ance

: Lon

gitu

dina

l stu

dies

on

the

pred

ictiv

e va

lue

of a

pos

itive

IGRA

are

stil

l em

ergi

ng in

TB

high

-inci

denc

e co

untr

ies

(≥ 10

0/10

0 00

0). I

t is

impo

rtan

t to

asse

ss w

heth

er IG

RA c

an b

e us

ed a

s a

repl

acem

ent f

or th

e w

idel

y us

ed T

ST.

Refe

renc

e st

anda

rd: A

ll di

agno

ses

of in

cide

nt a

ctiv

e TB

(mic

robi

olog

ical

ly c

onfir

med

or n

ot)

Stud

ies:

Any

long

itudi

nal s

tudy

des

ign

(e.g

. pro

spec

tive

or re

tros

pect

ive

coho

rt),

in T

B hi

gh-in

cide

nce

coun

trie

s, re

gard

less

of i

mm

unol

ogic

al s

tatu

s (e

.g. H

IV-in

fect

ed o

r not

) or B

CG s

tatu

s.

Ave

rage

follo

w-u

p sh

ould

be ≥

1 yea

r, bu

t can

be

eith

er a

ctiv

e or

pas

sive

.

Nos

of s

tudi

es a

nd

patie

nts

Des

ign

Qua

lity

Effec

tQ

ualit

y Im

port

ance

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

n Re

lativ

e (p

oole

d)

Abs

olut

e eff

ect

(GRA

DE)

A. S

R O

UTC

OM

E: P

ROG

RESS

ION

TO

ACT

IVE

TUBE

RCU

LOSI

S IN

UN

TREA

TED

IND

IVID

UA

LS

5 (N

= 7

675

for

TST,

N =

764

1 fo

r IG

RA) (

44-

48)

Pros

pect

ivel

y fo

llow

ed c

ohor

tsSe

rious

(A

1) (-

1)Se

rious

TS

T: I2 =

64.

4%

IGRA

: I2 =

49

.6%

(A2)

(-1)

Not

ser

ious

(A

3)

TST:

Ser

ious

im

prec

isio

nIG

RA: N

o se

rious

im

prec

isio

n (A

4) (-

1)

TST:

RR =

1.49

(9

5% C

I 0.7

9;2.

80)

I2 = 6

4.4%

IGRA

RR =

2.0

3(9

5% C

I 1.18

;3.5

0)I2 =

49.

6%

TST

10 m

ore

per 1

000

(4

few

er to

37

mor

e)IG

RA15

mor

e pe

r 100

0 (3

m

ore

to 3

6 m

ore)

Very

low

㊉◯

◯◯

Crit

ical

B. S

R O

UTC

OM

E (S

UB-

GRO

UP

AN

ALY

SIS)

: PRO

GRE

SSIO

N T

O A

CTIV

E TB

IN IM

MU

NO

COM

PRO

MIS

ED P

EOPL

E (H

IV A

ND

OTH

ER IM

MU

NO

SUPP

RESS

IVE

CON

DIT

ION

S)

2 (N

= 7

25 fo

r TS

T, N

= 7

10 fo

r IG

RA) (

44, 4

5)

Pros

pect

ivel

y fo

llow

ed c

ohor

t of

HIV

-infe

cted

w

omen

pre

- and

po

st-d

eliv

ery

of A

RTPr

ospe

ctiv

ely

follo

wed

coh

ort

of H

IV-in

fect

ed

indi

vidu

als

Serio

us

(B1)

(-1)

Serio

us

TST:

I2 = 7

7.4%

IGRA

: I2 =

78

.7%

(B2)

(-1)

Serio

us

(B3)

(-1)

Very

ser

ious

(B4)

(-2)

TS

T:RR

= 1.

64(9

5% C

I 0.2

4;11

.18)

IGRA

RR =

4.0

7 (9

5% C

I 0.

18;9

2.72

)

TST

39 m

ore

per 1

000

(46

few

er to

616

mor

e)IG

RA14

9 m

ore

per 1

000

(40

few

er to

443

8 m

ore)

Very

low

㊉◯

◯◯

Crit

ical

49

Nos

of s

tudi

es a

nd

patie

nts

Des

ign

Qua

lity

Effec

tQ

ualit

y Im

port

ance

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

n Re

lativ

e (p

oole

d)

Abs

olut

e eff

ect

(GRA

DE)

C. S

R O

UTC

OM

E (S

UB-

GRO

UP

AN

ALY

SIS)

: PRO

GRE

SSIO

N T

O A

CTIV

E TB

AM

ON

G C

ON

TACT

S O

F TB

CA

SES

1 (N

= 15

11 fo

r TS

T, N

= 14

98

for I

GRA

) (48

)

Pros

pect

ive

follo

w-

upSe

rious

(C1)

(-1)

Not

ass

esse

d;

sing

le s

tudy

(C2)

Serio

us

C3

(-1)

Serio

us

C4

(-1)

TST

RR, s

ingl

e st

udy

=1.3

1 (95

% C

I: 0.

85;2

.04)

IGRA

RR

, sin

gle

stud

y =

1.87

(95%

CI:

1.12;

3.11

)

TST

14 m

ore

per 1

000

(7

few

er to

45

mor

e)IG

RA28

mor

e pe

r 100

0

(4 m

ore

to 6

9 m

ore)

Very

low

㊉◯

◯◯

Crit

ical

D. S

R O

UTC

OM

E (S

UB-

GRO

UP

AN

ALY

SIS)

: PRO

GRE

SSIO

N T

O A

CTIV

E TB

AM

ON

G T

B H

EALT

H-C

ARE

WO

RKER

S

1 (N

= 19

5 fo

r TS

T, N

= 18

9 fo

r IG

RA) (

47)

Pros

pect

ive

follo

w-

upSe

rious

risk

of

bia

s (D

1) (-

1)

Not

ass

esse

d;

sing

le s

tudy

.

(D2)

Serio

us

D3

(-1)

Very

ser

ious

D

4 (-2

)TS

TRR

, sin

gle

stud

y =

0.40

(95%

CI:

0.02

;9.8

1)IG

RARR

, sin

gle

stud

y =

3.10

(95%

CI:

0.13

;75.

04)

TST

6 fe

wer

per

1000

(9

few

er to

82

mor

e)IG

RA(A

diff

eren

ce c

anno

t be

com

pute

d)

Very

low

㊉◯

◯◯

Crit

ical

E. S

R O

UTC

OM

E (S

UB-

GRO

UP

AN

ALY

SIS)

: PRO

GRE

SSIO

N T

O A

CTIV

E TB

AM

ON

G A

DO

LESC

ENTS

IN A

HIG

H-I

NCI

DEN

CE S

ETTI

NG

1 (N

= 5

244

for

both

test

s) (4

6)Pr

ospe

ctiv

e fo

llow

-up

Se

rious

(E

1) (-

1)N

ot a

sses

sed;

si

ngle

stu

dy

(E2)

Serio

us

E3 (-

1)N

o se

rious

E4

TS

TRR

, sin

gle

stud

y =

2.71

(95%

CI:

1.42;

5.15

)IG

RARR

, sin

gle

stud

y =

2.89

(95%

CI:

1.55;

5.41

)

TST

9 m

ore

per 1

000

(2

mor

e to

21 m

ore)

IGRA

10 m

ore

per 1

000

(3

mor

e to

22

mor

e)

Very

low

㊉◯

◯◯

Crit

ical



Not

es o

n G

RAD

E su

mm

ary

tabl

eO

vera

ll qu

ality

: A

ll st

udie

s st

art w

ith o

ne p

oint

doc

ked

off b

ecau

se n

one

wer

e RC

Ts. T

he lo

wes

t qua

lity

scor

e ac

hiev

able

is 1

out o

f 4; n

o m

inus

sco

res

are

give

n.Q

ualit

y as

sess

men

t: Ba

sed

on th

e re

lativ

e eff

ect m

easu

re (R

R or

IRR)

for b

oth

TST

and

IGRA

. Stu

dies

not

mar

ked

dow

n if

estim

ates

for b

oth

test

s sc

ore

high

on

a sp

ecifi

c G

RAD

E qu

ality

item

.

Oth

er s

tudy

qua

lity

cons

ider

atio

ns: N

ewca

stle

-Ott

awa

Scal

e qu

ality

item

s w

ere

cons

ider

ed w

hen

asse

ssin

g th

e ris

k of

bia

s. O

ne p

oint

will

be

dock

ed if

at l

east

one

con

cern

is p

rese

nt.

A1:

Risk

of b

ias

is p

ossi

ble.

Issu

es in

the

stud

ies

incl

ude

sele

ctio

n bi

as, r

isk

of in

corp

orat

ion

bias

, asc

erta

inm

ent a

nd p

ublic

atio

n bi

as. M

etho

ds fo

r asc

erta

inin

g TB

incl

uded

mic

robi

olog

ical

met

hods

, but

not

all

inci

dent

TB

case

s ha

d a

defin

ite c

ultu

re-c

onfir

med

dia

gnos

is o

f TB.

Pub

licat

ion

bias

not

form

ally

ass

esse

d, b

ut e

xpec

ted

to b

e lik

ely.

Sev

eral

larg

e pr

ospe

ctiv

e st

udie

s are

ong

oing

and

/or u

npub

lishe

d; th

eir r

esul

ts w

ere

not i

nclu

ded

in th

is a

naly

sis.

H

owev

er, a

dditi

on o

f res

ults

is n

ot e

xpec

ted

to c

hang

e th

e ov

eral

l con

clus

ions

of t

his

revi

ew.

A2:

Ser

ious

une

xpla

ined

inco

nsis

tenc

y of

RR

estim

ate

for T

ST. P

oint

s do

cked

if s

erio

us in

cons

iste

ncy

iden

tified

in e

ither

est

imat

e.A

3: A

lthou

gh th

e nu

mbe

r of s

tudi

es in

clud

ed is

sm

all,

they

invo

lve

a ra

nge

of p

opul

atio

ns, i

nclu

ding

adu

lts a

nd c

hild

ren,

imm

unoc

ompr

omis

ed p

eopl

e an

d TB

con

tact

s, p

rovi

ding

dire

ct e

vide

nce

for t

hese

gro

ups.

A4:

Ser

ious

impr

ecis

ion

of R

R es

timat

e fo

r TST

. Low

er li

mit

of 9

5% C

I ind

icat

es la

ck o

f pre

dict

ive

utili

ty. P

oint

s do

cked

if s

erio

us im

prec

isio

n id

entifi

ed in

eith

er e

stim

ate.

B1: R

isk

of b

ias

is p

ossi

ble.

Issu

es in

clud

e se

lect

ion

bias

, ris

k of

inco

rpor

atio

n bi

as, a

scer

tain

men

t and

pub

licat

ion

bias

. Inc

orpo

ratio

n bi

as c

ould

not

be

rule

d ou

t in

the

coho

rt th

at in

clud

ed a

ntep

artu

m a

nd p

ostp

artu

m w

omen

be

caus

e in

form

atio

n w

as n

ot a

vaila

ble;

mor

eove

r, th

ere

are

conc

erns

with

sel

ectio

n. T

he A

RT c

ohor

t stu

dy re

port

ed re

fere

nce

stan

dard

s th

at d

o no

t acc

ount

for i

ndex

test

s; h

owev

er, a

sses

sors

wer

e no

t blin

ded

to b

asel

ine

TST

resu

lts th

at w

ere

reco

rded

in p

atie

nt re

cord

s. M

etho

ds fo

r asc

erta

inin

g TB

incl

uded

mic

robi

olog

ical

met

hods

, but

not

all

inci

dent

TB

case

s ha

d a

defin

ite d

iagn

osis

of T

B. P

ublic

atio

n bi

as n

ot fo

rmal

ly a

sses

sed,

but

exp

ecte

d to

be

like

ly. S

ever

al la

rge

pros

pect

ive

stud

ies

are

ongo

ing

and/

or a

re u

npub

lishe

d; th

eir r

esul

ts w

ere

not i

nclu

ded

in th

is a

naly

sis.

How

ever

, add

ition

of r

esul

ts is

not

exp

ecte

d to

cha

nge

the

over

all c

oncl

usio

ns o

f thi

s re

view

.B2

: Ser

ious

une

xpla

ined

inco

nsis

tenc

y in

RR

estim

ates

for b

oth

TST

and

IGRA

. B3

: Thi

s po

oled

est

imat

e is

bas

ed o

n on

ly tw

o st

udie

s: o

ne s

tudy

of H

IV-in

fect

ed p

eopl

e on

ART

with

a m

edia

n C

D4+

app

roxi

mat

ely

250,

and

one

on

HIV

-infe

cted

ant

epar

tum

and

pos

tpar

tum

wom

en. N

o di

rect

evi

denc

e fo

r tr

eatm

ent-

naïv

e pa

tient

s an

d/or

HIV

-infe

cted

pat

ient

s w

ith h

igh

CD

4 co

unts

or o

ther

sub

-pop

ulat

ions

of H

IV-in

fect

ed in

divi

dual

s (e

.g. c

hild

ren)

. B4

: Ver

y se

rious

impr

ecis

ion

of R

R es

timat

es fo

r bot

h TS

T an

d IG

RA. C

Is a

re w

ide

and

indi

cate

bot

h si

gnifi

cant

pre

dict

ive

perf

orm

ance

and

lack

of p

redi

ctiv

e ut

ility

. Stu

dies

had

few

eve

nts.

C1:

Risk

of b

ias

is p

ossi

ble.

Issu

es in

clud

e se

lect

ion

bias

, ris

k of

inco

rpor

atio

n bi

as (

no in

form

atio

n) a

nd p

ublic

atio

n bi

as. P

ublic

atio

n bi

as n

ot fo

rmal

ly a

sses

sed,

but

exp

ecte

d to

be

likel

y. S

ever

al la

rge

pros

pect

ive

stud

ies

are

ongo

ing

and/

or a

re u

npub

lishe

d; th

eir r

esul

ts w

ere

not i

nclu

ded

in th

is a

naly

sis.

How

ever

, add

ition

of r

esul

ts is

not

exp

ecte

d to

cha

nge

the

over

all c

oncl

usio

ns o

f thi

s re

view

.C

2: In

cons

iste

ncy

not a

sses

sed.

C3:

Thi

s si

ngle

stu

dy c

ompr

ises

hou

seho

ld c

ase

cont

acts

in a

hig

h-in

cide

nce

coun

try.

No

dire

ct e

vide

nce

for o

ther

sub

popu

latio

ns o

f cas

e co

ntac

ts.

C4:

Ser

ious

impr

ecis

ion

of T

ST e

ffect

est

imat

es. L

ower

lim

it of

95%

CI i

ndic

ates

lack

of p

redi

ctiv

e ut

ility

.

D1:

Risk

of b

ias

is p

ossi

ble.

Issu

es in

clud

e se

lect

ion

bias

, lac

k of

use

of m

icro

biol

ogic

al to

ols

in m

etho

ds to

asc

erta

in T

B, in

corp

orat

ion

bias

and

pub

licat

ion

bias

. Pub

licat

ion

bias

not

form

ally

ass

esse

d, b

ut e

xpec

ted

to b

e lik

ely.

Se

vera

l lar

ge p

rosp

ectiv

e st

udie

s ar

e on

goin

g an

d/or

are

unp

ublis

hed;

thei

r res

ults

wer

e no

t inc

lude

d in

this

ana

lysi

s. H

owev

er, a

dditi

on o

f res

ults

is n

ot e

xpec

ted

to c

hang

e th

e ov

eral

l con

clus

ions

of t

his

revi

ew.

D2:

Inco

nsis

tenc

y no

t ass

esse

d.D

3: T

his

sing

le s

tudy

com

pris

es h

ealth

car

e w

orke

rs a

t a p

rimar

y he

alth

car

e cl

inic

. No

dire

ct e

vide

nce

for o

ther

sub

popu

latio

ns o

f hea

lth c

are

wor

kers

or a

ll se

ttin

gs o

f hea

lth c

are.

D

4: V

ery

serio

us im

prec

isio

n of

IGRA

and

TST

effe

ct e

stim

ates

; CIs

are

wid

e an

d in

dica

te b

oth

sign

ifica

nt p

redi

ctiv

e pe

rfor

man

ce a

nd la

ck o

f pre

dict

ive

utili

ty.

E1: R

isk

of b

ias

is p

ossi

ble.

Issu

es in

clud

e se

lect

ion

bias

, inc

orpo

ratio

n of

inde

x te

sts

in m

etho

ds to

asc

erta

in in

cide

nt T

B an

d pu

blic

atio

n bi

as. P

ublic

atio

n bi

as n

ot fo

rmal

ly a

sses

sed,

but

exp

ecte

d to

be

likel

y. S

ever

al la

rge

pros

pect

ive

stud

ies

are

ongo

ing

and/

or a

re u

npub

lishe

d; th

eir r

esul

ts w

ere

not i

nclu

ded

in th

is a

naly

sis.

How

ever

, add

ition

of r

esul

ts is

not

exp

ecte

d to

cha

nge

the

over

all c

oncl

usio

ns o

f thi

s re

view

. E2

: Inc

onsi

sten

cy n

ot a

sses

sed.

E3

: Thi

s si

ngle

stu

dy c

ompr

ises

ado

lesc

ents

in a

hig

h-in

cide

nce

sett

ing.

No

dire

ct e

vide

nce

for o

ther

sub

popu

latio

ns o

f chi

ldre

n or

ado

lesc

ents

. E4

: No

serio

us im

prec

isio

n: F

ew e

vent

s w

ith la

rge

sam

ple

size

.

51

PICO

5: S

houl

d 3-

mon

th d

aily

rifa

mpi

cin

plus

ison

iazi

d (3

RH) b

e off

ered

as

a pr

even

tive

trea

tmen

t opt

ion

for c

hild

ren

and

adol

esce

nts

<15

year

s of

age

as

an a

ltern

ativ

e to

6 o

r 9 m

onth

s is

onia

zid

(IN

H) m

onot

hera

py in

hig

h TB

inci

denc

e co

untr

ies?

3-m

onth

dai

ly ri

fam

pici

n an

d is

onia

zid

in c

hild

ren

and

adol

esce

nts

< 15

yea

rs

Ove

rall

qual

ity: l

ow

Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3–4-

mon

th

daily

rif

ampi

cin

and

ison

iazi

d

6–9-

mon

th

ison

iazi

d m

onot

hera

py

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

“RA

DIO

LOG

ICA

L” T

B D

ISEA

SE: F

OLL

OW

UP:

RA

NG

E 3–

7 Y

EARS

TO

7–1

1 YEA

RS; A

SSES

SED

WIT

H: C

HES

T RA

DIO

GRA

PHY

1 (49

)Ra

ndom

ised

tr

ial

Serio

us1

Not

ser

ious

Se

rious

2N

ot s

erio

us

Non

e 26

/220

(1

1.8%

) 48

/200

(2

4.0%

) RR

0.4

92

(0.3

18;0

.762

) 12

2 fe

wer

pe

r 100

0 (f

rom

57

few

er to

164

few

er)

㊉㊉

◯◯

Lo

w

Crit

ical

MO

RTA

LITY

0 C

anno

t be

estim

ated

–

Impo

rtan

t

AD

VER

SE E

VEN

TS: F

OLL

OW

UP:

RA

NG

E 3–

7 Y

EARS

TO

7–1

1 YEA

RS; A

SSES

SED

BY:

REC

OG

NIT

ION

OF

SYM

PTO

MS

AN

D E

LEVA

TED

LIV

ER E

NZY

MES

1 (49

)Ra

ndom

ized

tr

ial

Very

se

rious

1,3

Not

ser

ious

Se

rious

4N

ot s

erio

us

Non

e 27

/650

(4

.2%

) 25

/200

(1

2.5%

) RR

0.3

32

(0.19

7;0.

559)

83

few

er

per 1

000

(fro

m 5

5 fe

wer

to

100

few

er)

㊉◯

◯◯

Ve

ry lo

w

Crit

ical

AD

VER

SE E

VEN

TS: F

OLL

OW

UP:

MED

IAN

97–

197

DAY

S; A

SSES

SED

BY:

LIV

ER T

OX

ICIT

Y T

EST

AN

D C

LIN

ICA

L

1 (50

)O

bser

vatio

nal

stud

y Se

rious

5N

ot s

erio

us

Serio

us4

Serio

us6

Non

e 1/

220

(0.5

%)

5/26

4 (1

.9%

) RR

0.2

4 (0

.03;

2.04

) 14

few

er p

er

1000

(f

rom

18

few

er to

20

mor

e)

㊉◯

◯◯

Ve

ry lo

w

Crit

ical

COM

PLET

ION

RAT

E: F

OLL

OW

UP:

RA

NG

E 3–

7 Y

EARS

TO

7–1

1 YEA

RS#

1 (49

)Ra

ndom

ized

tr

ial

Serio

us7

Not

ser

ious

Se

rious

4N

ot s

erio

us

Non

e 22

0/23

8 (9

2.4%

) 20

0/23

2 (8

6.2%

) RR

1.07

(1

.01;1

.14)

60 m

ore

per 1

000

(fro

m 9

m

ore

to 12

1 m

ore)

㊉㊉

◯◯

Lo

w

Crit

ical



Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3–4-

mon

th

daily

rif

ampi

cin

and

ison

iazi

d

6–9-

mon

th

ison

iazi

d m

onot

hera

py

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

COM

PLET

ION

RAT

E: A

SSES

SED

BY:

CO

MPL

ETIO

N O

F >

80%

OF

TREA

TMEN

T W

ITH

OU

T IN

TERR

UPT

ION

OF

> 2

MO

NTH

S

1 (51

)O

bser

vatio

nal

stud

y Se

rious

5N

ot s

erio

us

Not

ser

ious

Se

rious

8 N

one

48/7

2 (6

6.7%

) 29

/105

(2

7.6%

) RR

2.4

1 (1

.70;

3.43

) 38

9 m

ore

per 1

000

(fro

m 19

3 m

ore

to 6

71

mor

e)

㊉◯

◯◯

Ve

ry lo

w

Crit

ical

DRU

G-R

ESIS

TAN

T TB

0 C

anno

t be

estim

ated

–

Impo

rtan

t

From

refe

renc

es (4

9–51

)1

Alth

ough

ther

e w

as a

risk

of s

elec

tion

bias

, the

cha

ract

eris

tics

of th

e tw

o gr

oups

wer

e si

mila

r. Pa

tient

s w

ith p

oor c

ompl

ianc

e w

ere

not i

nclu

ded

in th

e an

alys

is o

f tre

atm

ent o

utco

mes

. Dow

ngra

ded

by o

ne le

vel.

2 Th

ere

was

no

clin

ical

dis

ease

. The

out

com

e re

port

ed w

as n

ew ra

diog

raph

ic fi

ndin

gs s

ugge

stin

g po

ssib

le a

ctiv

e di

seas

e. N

o da

ta c

ompa

red

with

6H

. Dow

ngra

ded

by o

ne le

vel.

3 A

hig

h ris

k of

det

ectio

n bi

as d

ue to

lack

of b

lindi

ng. T

he R

H g

roup

incl

uded

par

ticip

ants

enr

olle

d du

ring

the

seco

nd p

erio

d, w

hose

cha

ract

eris

tics

wer

e di

ffere

nt; t

hey

wer

e no

t ran

dom

ized

bet

wee

n th

e RH

gro

up a

nd th

e 9H

gr

oup.

Dow

ngra

ded

by tw

o le

vels

.4

No

data

com

pare

d w

ith 6

H. D

owng

rade

d by

one

leve

l.5

Risk

of b

ias

due

to p

oor c

ompa

rabi

lity

of th

e tw

o gr

oups

. Dow

ngra

ded

by o

ne le

vel.

6 Lo

w e

vent

rate

and

wid

e 95

% C

I. D

owng

rade

d by

one

leve

l.7

Lack

of b

lindi

ng. M

edic

atio

n ad

here

nce

test

was

per

form

ed a

t hom

e by

par

ents

. Alth

ough

ther

e w

as a

risk

of s

elec

tion

bias

, the

cha

ract

eris

tics

of th

e tw

o gr

oups

wer

e si

mila

r. D

owng

rade

d by

one

leve

l.8

Wid

e 95

% C

I. D

owng

rade

d by

one

leve

l.#

The

stu

dy re

port

ed a

dher

ence

rate

s; c

ompl

ianc

e w

as c

onsi

dere

d to

be

poor

if n

o m

edic

atio

n w

as d

etec

ted

in u

rine

strip

s or i

f pat

ient

s did

not

retu

rn fo

r fol

low

-up

visi

ts o

r wer

e lo

st to

follo

w-u

p. P

oor c

ompl

ianc

e w

as c

onsi

dere

d no

n-co

mpl

etio

n in

the

anal

ysis

.

53

PICO

6: S

houl

d 3-

mon

th w

eekl

y rif

apen

tine

and

ison

iazi

d be

off

ered

as

an a

ltern

ativ

e re

gim

en to

ison

iazi

d m

onot

hera

py fo

r tr

eatm

ent o

f LTB

I in

high

TB

inci

denc

e co

untr

ies?

3-m

onth

wee

kly

rifap

entin

e pl

us is

onia

zid

or d

aily

ison

iazi

d m

onot

hera

py fo

r LTB

I tre

atm

ent i

n ad

ults

with

HIV

Po

pula

tion:

Adu

lts w

ith H

IVCo

mpa

rison

: 6 o

r 9 m

onth

s of

ison

iazi

d m

onot

hera

pyO

vera

ll qu

ality

: hig

h

Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3 m

onth

s w

eekl

y rif

apen

tine

+ is

onia

zid

6 or

9

mon

ths

ison

iazi

d

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

ACT

IVE

TB

2 (5

2,53

)RC

Ts

Not

se

rious

N

ot s

erio

us

Not

ser

ious

1Se

rious

2N

one

26/5

34

(4.9

%)

28/5

20

(5.4

%)

RR 0

.733

(0

.234

;2.2

95)

14 fe

wer

per

10

00

(fro

m 4

1 few

er

to 7

0 m

ore)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

ALL

-CA

USE

MO

RTA

LITY

2 (5

2,53

)RC

TsN

ot

serio

us

Not

ser

ious

N

ot s

erio

us1

Serio

us2

Non

e 23

/535

(4

.3%

) 30

/513

(5

.8%

) RR

0.7

46

(0.4

38;1.

270)

15

few

er p

er

1000

(f

rom

16 m

ore

to 3

3 fe

wer

)

㊉㊉

㊉◯

M

oder

ate

Impo

rtan

t

AN

Y A

DV

ERSE

EV

ENTS

(GRA

DE

III O

R IV

)

2 (5

2,53

)RC

TsSe

rious

3N

ot s

erio

us

Not

ser

ious

1N

ot s

erio

us

Non

e39

/535

(7

.3%

) 59

/513

(1

1.5%

) RR

0.6

27

(0.4

26;0

.921

) 43

few

er p

er

1000

(f

rom

9 fe

wer

to

66

few

er)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

HEP

ATO

TOX

ICIT

Y

2 (5

2,53

)RC

TsN

ot

serio

us4

Not

ser

ious

N

ot s

erio

us1

Not

ser

ious

N

one

8/53

5 (1

.5%

) 30

/513

(5

.8%

) RR

0.2

56

(0.11

8;0.

553)

44

few

er p

er

1000

(f

rom

26

few

er to

52

few

er)

㊉㊉

㊉㊉

H

igh

Crit

ical



Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3 m

onth

s w

eekl

y rif

apen

tine

+ is

onia

zid

6 or

9

mon

ths

ison

iazi

d

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

DRU

G R

ESIS

TAN

T TB

2 (5

2,53

)RC

Ts

Not

se

rious

N

ot s

erio

us

Not

ser

ious

1Ve

ry

serio

us5

Non

e3/

534

(0.6

%)

1/52

0 (0

.2%

) RR

2.0

01

(0.2

59;15

.436

) 2

mor

e pe

r 10

00

(fro

m 1

few

er

to 2

8 m

ore)

㊉㊉

◯◯

Lo

w

Impo

rtan

t

COM

PLET

ION

RAT

E

2 (5

2,53

)RC

TsN

ot

serio

us

Not

ser

ious

N

ot s

erio

us1

Not

ser

ious

N

one

497/

534

(93.

1%)

397/

520

(76.

3%)

RR 1.

255

(1.0

14;1.

553)

19

5 m

ore

per

1000

(f

rom

11 m

ore

to 4

22 m

ore)

㊉㊉

㊉㊉

H

igh

Crit

ical

1 A

lthou

gh o

ne o

f the

tria

ls w

as c

ondu

cted

in lo

w T

B in

cide

nce

coun

trie

s, th

is is

unl

ikel

y to

affe

ct th

e re

lativ

e eff

ect o

f rifa

pent

ine

+ is

onia

zid

com

pare

d w

ith is

onia

zid

mon

othe

rapy

. Not

dow

ngra

ded.

2

95%

CIs

of b

oth

rela

tive

and

abso

lute

effe

ct in

clud

e ap

prec

iabl

e be

nefit

and

har

m w

ith 3

HP.

3

Both

tria

ls w

ere

open

-labe

l, w

hich

may

hav

e in

trod

uced

bia

s in

asc

erta

inm

ent o

f adv

erse

eve

nts.

4

Alth

ough

the

tria

ls w

ere

open

-labe

l, th

is is

unl

ikel

y to

affe

ct d

etec

tion

of h

epat

otox

icity

, whi

ch is

usu

ally

don

e by

obj

ectiv

e m

easu

rem

ent (

i.e. b

lood

test

s.).

Not

dow

ngra

ded.

5

Ver

y lo

w e

vent

rate

s. U

pper

lim

it of

95%

CI o

f bot

h re

lativ

e an

d ab

solu

te e

ffect

incl

ude

appr

ecia

ble

harm

with

3H

P. D

owng

rade

d by

two

leve

ls.

55

3-m

onth

wee

kly

rifap

entin

e pl

us is

onia

zid

or d

aily

ison

iazi

d m

onot

hera

py fo

r tre

atm

ent o

f LTB

I in

adul

ts w

ithou

t HIV

Po

pula

tion:

Adu

lts w

ithou

t HIV

Com

paris

on: 6

or 9

mon

ths

of is

onia

zid

mon

othe

rapy

Ove

rall

qual

ity: m

oder

ate

Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3-m

onth

rif

apen

tine

+ is

onia

zid

6 or

9 m

onth

s is

onia

zid

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

ACT

IVE

TB

1 (54

)RC

TN

ot s

erio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us2

Non

e 7/

3986

(0

.2%

) 15

/374

5 (0

.4%

) RR

0.4

38

(0.17

9;1.0

74)

2 fe

wer

per

10

00

(fro

m 0

few

er

to 3

few

er)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

ALL

-CA

USE

MO

RTA

LITY

1 (54

)RC

TN

ot s

erio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us3

Non

e31

/398

6 (0

.8%

) 39

/375

9 (1

.0%

) RR

0.7

40

(0.4

62;1.

183)

3

few

er p

er

1000

(f

rom

2 m

ore

to 6

few

er)

㊉㊉

㊉◯

M

oder

ate

Impo

rtan

t

AN

Y A

DV

ERSE

EV

ENTS

(GRA

DE

III O

R IV

)

1 (54

)RC

TSe

rious

4N

ot s

erio

us

Serio

us1

Not

ser

ious

N

one

229/

4040

(5

.7%

) 24

4/37

59

(6.5

%)

RR 0

.873

(0

.733

;1.04

0)

8 fe

wer

per

10

00

(fro

m 3

mor

e to

17 fe

wer

)

㊉㊉

◯◯

Lo

w

Crit

ical

HEP

ATO

TOX

ICIT

Y

1 (54

)RC

TN

ot

serio

us5

Not

ser

ious

Se

rious

1N

ot s

erio

us

Non

e18

/404

0 (0

.4%

) 10

3/37

59

(2.7

%)

RR 0

.163

(0.0

99;0

.268

) 23

few

er p

er

1000

(f

rom

20

few

er

to 2

5 fe

wer

)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

DRU

G-R

ESIS

TAN

T TB

1 (54

)RC

TN

ot s

erio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us3

Non

e1/

3986

(0

.0%

) 2/

3745

(0

.1%)

RR 0

.470

(0

.043

;5.17

9)

0 fe

wer

per

10

00

(fro

m 1

few

er

to 2

mor

e)

㊉㊉

㊉◯

M

oder

ate

Impo

rtan

t



Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3-m

onth

rif

apen

tine

+ is

onia

zid

6 or

9 m

onth

s is

onia

zid

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

COM

PLET

ION

RAT

E

1 (54

)RC

TN

ot s

erio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us

Non

e32

73/3

985

(82.

1%)

2585

/374

5 (6

9.0%

) RR

1.19

0 (1

.159;

1.221

) 13

1 mor

e pe

r 10

00

(fro

m 11

0 m

ore

to 15

3 m

ore)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

1 N

o co

mpa

rison

with

6 m

onth

s of

ison

iazi

d. T

he s

tudy

incl

uded

2.7

% H

IV-p

ositi

ve p

artic

ipan

ts. A

lthou

gh th

e tr

ial w

as c

ondu

cted

in lo

w T

B in

cide

nce

coun

trie

s, th

is is

unl

ikel

y to

affe

ct re

lativ

e eff

ect o

f rifa

pent

ine

+ is

onia

zid

com

pare

d w

ith is

onia

zid

mon

othe

rapy

. Dow

ngra

ded

by o

ne le

vel.

2

Alth

ough

the

95%

CI o

f RR

is w

ide,

the

num

ber o

f eve

nts

was

sm

all a

nd th

e C

I of a

bsol

ute

effec

t is

narr

ow. T

he re

sult

also

met

pre

-sta

ted

non-

infe

riorit

y m

argi

n. N

ot d

owng

rade

d.

3

Alth

ough

the

95%

CI o

f RR

is w

ide,

the

num

ber o

f eve

nts

was

sm

all a

nd th

e C

I of a

bsol

ute

effec

t is

narr

ow. N

ot d

owng

rade

d.

4

An

open

-labe

l des

ign

of th

e tr

ial m

ay h

ave

intr

oduc

ed a

scer

tain

men

t bia

s.

5

Alth

ough

the

tria

l was

ope

n-la

bel,

this

is u

nlik

ely

to a

ffect

det

ectio

n of

hep

atot

oxic

ity, w

hich

is u

sual

ly d

one

by o

bjec

tive

mea

sure

men

t (i.e

. blo

od te

sts)

. Not

dow

ngra

ded.

57

3-m

onth

wee

kly

rifap

entin

e pl

us is

onia

zid

or d

aily

ison

iazi

d m

onot

hera

py fo

r tre

atm

ent o

f LTB

I in

child

ren

and

adol

esce

nts

Popu

latio

n: C

hild

ren

and

adol

esce

nts

Com

paris

on: 6

or 9

mon

ths

ison

iazi

dO

vera

ll qu

ality

: mod

erat

e

Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3-m

onth

rif

apen

tine

+ is

onia

zid

6 or

9

mon

ths

ison

iazi

d

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

ACT

IVE

TB

1 (55

)RC

TN

ot

serio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us2

Non

e 0/

471

(0.0

%)

3/43

4 (0

.7%

) RR

0.13

2 (0

.007

;2.5

42)

6 fe

wer

per

10

00

(fro

m 7

few

er to

11

mor

e)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

ALL

-CA

USE

MO

RTA

LITY

1 (55

)RC

TN

ot

serio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us3

Non

e0/

539

(0.0

%)

2/49

3 (0

.4%

) RR

0.18

3 (0

.009

;3.8

02)

3 fe

wer

per

1000

(f

rom

4 fe

wer

to

11 m

ore)

㊉㊉

㊉◯

M

oder

ate

Impo

rtan

t

AN

Y A

DV

ERSE

EV

ENTS

(GRA

DE

III O

R IV

)

1 (55

)RC

TSe

rious

4N

ot s

erio

us

Serio

us1

Not

ser

ious

3N

one

7/53

9 (1

.3%

) 8/

493

(1.6

%)

RR 0

.875

(0

.320

;2.3

96)

2 fe

wer

per

10

00

(fro

m 11

few

er to

23

mor

e)

㊉㊉

◯◯

Lo

w

Crit

ical

HEP

ATO

TOX

ICIT

Y

1 (55

)RC

TN

ot

serio

us5

Not

ser

ious

Se

rious

1N

ot s

erio

us

Non

e0/

539

(0.0

%)

0/49

3 (0

.0%

) C

anno

t be

estim

ated

0

few

er p

er

1000

(f

rom

4 fe

wer

to

4 m

ore)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

DRU

G-R

ESIS

TAN

T TU

BERC

ULO

SIS

0C

anno

t be

estim

ated

–

Impo

rtan

t



Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

t

Qua

lity

Impo

rtan

ceN

o. o

f st

udie

sSt

udy

desi

gnRi

sk o

f bia

sIn

cons

iste

ncy

Indi

rect

ness

Impr

ecis

ion

Oth

er

cons

ider

atio

ns

3-m

onth

rif

apen

tine

+ is

onia

zid

6 or

9

mon

ths

ison

iazi

d

Rela

tive

(95%

CI)

Abs

olut

e (9

5% C

I)

COM

PLET

ION

RAT

E

1 (55

)RC

TN

ot

serio

us

Not

ser

ious

Se

rious

1N

ot s

erio

us

Non

e 41

5/47

1 (8

8.1%

) 35

1/43

4 (8

0.9%

) RR

1.08

9 (1

.030

;1.15

3)

72 m

ore

per

1000

(f

rom

24

mor

e to

124

mor

e)

㊉㊉

㊉◯

M

oder

ate

Crit

ical

1 N

o co

mpa

rison

aga

inst

6 m

onth

s of

ison

iazi

d. A

lthou

gh t

he t

rial w

as c

ondu

cted

in lo

w T

B in

cide

nce

coun

trie

s, t

his

is u

nlik

ely

to a

ffect

rel

ativ

e eff

ect

of r

ifape

ntin

e +

ison

iazi

d co

mpa

red

with

ison

iazi

d m

onot

hera

py.

Dow

ngra

ded

by o

ne le

vel.

2 A

lthou

gh th

e 95

% C

I of t

he R

R is

wid

e, th

e nu

mbe

r of e

vent

s w

as s

mal

l and

the

CI o

f abs

olut

e eff

ect i

s na

rrow

. The

resu

lt al

so m

et p

re-s

tate

d no

n-in

ferio

rity

mar

gin.

Not

dow

ngra

ded.

3

Alth

ough

the

95%

CI o

f the

RR

is w

ide,

the

num

ber o

f eve

nts

was

sm

all a

nd th

e C

I of a

bsol

ute

effec

t is

narr

ow. N

ot d

owng

rade

d.

4 A

n op

en-la

bel d

esig

n of

the

tria

l may

hav

e in

trod

uced

asc

erta

inm

ent b

ias.

5

Alth

ough

the

tria

l was

ope

n-la

bel,

this

is u

nlik

ely

to a

ffect

det

ectio

n of

hep

atot

oxic

ity, w

hich

is u

sual

ly d

one

by o

bjec

tive

mea

sure

men

t (i.e

. blo

od te

sts)

. Not

dow

ngra

ded.

59

PICO

7: S

houl

d pr

even

tive

trea

tmen

t be

reco

mm

ende

d fo

r con

tact

s of

pat

ient

s w

ith m

ultid

rug-

resi

stan

t or

rifam

pici

n-re

sist

ant T

B?Pr

even

tive

trea

tmen

t for

con

tact

s of

pat

ient

s w

ith m

ultid

rug-

or r

ifam

pici

n-re

sist

ant T

BFi

ve s

tudi

es th

at in

clud

ed fe

wer

than

20

part

icip

ants

who

com

plet

ed p

reve

ntiv

e TB

trea

tmen

t wer

e ex

clud

ed. I

n ad

ditio

n, th

e st

udy

by K

risti

was

exc

lude

d as

onl

y is

onia

zid

mon

othe

rapy

was

gi

ven.

Ove

rall

qual

ity: v

ery

low

Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

tQ

ualit

yIm

port

ance

No.

of

stud

ies

Stud

y de

sign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nO

ther

co

nsid

erat

ions

Prev

entiv

e tr

eatm

ent

No

trea

tmen

tRe

lativ

e (9

5% C

I)A

bsol

ute

(95%

CI)

INCI

DEN

CE O

F A

CTIV

E TB

DIS

EASE

(BO

TH D

RUG

-SU

SCEP

TIBL

E A

ND

DRU

G-R

ESIS

TAN

T TB

)

4 (5

6–59

)O

bser

vatio

nal

Very

se

rious

1N

ot s

erio

us

Not

ser

ious

Ve

ry

serio

us3

Non

e2/

41

(4.9

%)

13/6

4 (2

0.3%

)0.

20

(0.0

4;0.

94)4

154

few

er p

er

1000

(273

few

er

to 3

6 fe

wer

)

㊉◯

◯◯

Ve

ry lo

w

Crit

ical

0/93

(0

%)

3/15

(2

0%)

0.02

(0

.00;

0.39

)520

0 fe

wer

per

10

00 (4

03 fe

wer

to

3 m

ore)

0/21

(0

%)

0/10

(0

%)

–60

mor

e pe

r 100

0 (1

38 fe

wer

to 13

8 m

ore)

0/30

(0

%)

0/16

6 (0

%)

–70

mor

e pe

r 100

0 (4

5 fe

wer

to 4

5 m

ore)

INCI

DEN

CE O

F M

DR-

TB

32

(56,

57,

59

)

Obs

erva

tiona

lVe

ry

serio

us1

Not

ser

ious

Not

ser

ious

Ve

ry

serio

us3

Non

e 0/

93

(0%

)3/

15

(20%

)0.

02

(0.0

0;0.

39)5

200

few

er p

er

1000

(403

few

er

to 3

mor

e)

㊉◯

◯◯

Ve

ry lo

w

Crit

ical

0/21

(0

%)

0/10

(0

%)

–60

mor

e pe

r 100

0 (1

38 fe

wer

to 13

8 m

ore)

0/30

(0

%)

0/16

6 (0

%)

–70

mor

e pe

r 100

0 (4

5 fe

wer

to 4

5 m

ore)



Qua

lity

asse

ssm

ent

No.

of p

atie

nts

Effec

tQ

ualit

yIm

port

ance

No.

of

stud

ies

Stud

y de

sign

Risk

of b

ias

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nO

ther

co

nsid

erat

ions

Prev

entiv

e tr

eatm

ent

No

trea

tmen

tRe

lativ

e (9

5% C

I)A

bsol

ute

(95%

CI)

MO

RTA

LITY

0N

o ev

iden

ce

Can

not b

e es

timat

ed

– Im

port

ant

AD

VER

SE E

VEN

TS

0N

o ev

iden

ce

Can

not b

e es

timat

ed

– C

ritic

al

DEV

ELO

PMEN

T O

F D

RUG

RES

ISTA

NCE

0N

o ev

iden

ce

Impo

rtan

t

1 Ri

sk o

f bia

s in

sel

ectio

n of

the

cont

rol g

roup

, and

non

e of

the

stud

ies

adju

sted

for c

onfo

unde

rs. D

owng

rade

d by

two

leve

ls.

2

The

stud

y by

Sha

af e

t al.

was

exc

lude

d, a

s th

e in

cide

nce

of M

DR-

TB w

as n

ot re

port

ed.

3

Smal

l sam

ple

size

s an

d w

ide

95%

CIs

. Dow

ngra

ded

by tw

o le

vels

.4

Refe

renc

e (5

8)5

Refe

renc

e (5

6)6

Refe

renc

e (5

7)7

Refe

renc

e (5

9)

61

References

1. Kasambira TS, Shah M, Adrian PV, Holshouser M, Madhi SA, Chaisson RE, et al. QuantiFERON–TB Gold In–Tube for the detection of Mycobacterium tuberculosis infection in children with household tuberculosis contact. Int J Tuberc Lung Dis. 2011;15(5):628–34.

2. Kenyon TA, Creek T, Laserson K, Makhoa M, Chimidza N, Mwasekaga M, et al. Risk factors for transmission of Mycobacterium tuberculosis from HIV–infected tuberculosis patients, Botswana. Int J Tuberc Lung Dis. 2002;6(10):843–50.

3. Klausner JD, Ryder RW, Baende E, Lelo U, Williame JC, Ngamboli K, et al. Mycobacterium tuberculosis in household contacts of human immunodeficiency virus type 1–seropositive patients with active pulmonary tuberculosis in Kinshasa, Zaire. J Infect Dis. 1993;168(1):106–11.

4. Bokhari SY, Ahmad A, Shaikh MY, Ahmad I. A study of tuberculosis contacts. J Pak Med Assoc. 1987;37(2):48–52.

5. Biraro IA, Kimuda S, Egesa M, Cose S, Webb EL, Joloba M, et al. The Use of Interferon Gamma Inducible Protein 10 as a Potential Biomarker in the Diagnosis of Latent Tuberculosis Infection in Uganda. PLoS One. 2016;11(1):e0146098.

6. Rutherford ME, Nataprawira M, Yulita I, Apriani L, Maharani W, van Crevel R, et al. QuantiFERON(R)–TB Gold In–Tube assay vs. tuberculin skin test in Indonesian children living with a tuberculosis case. Int J Tuberc Lung Dis. 2012;16(4):496–502.

7. Tornee S, Kaewkungwal J, Fungladda W, Silachamroon U, Akarasewi P, Sunakorn P. Risk factors for tuberculosis infection among household contacts in Bangkok, Thailand. The Southeast Asian journal of tropical medicine and public health. 2004;35(2):375–83.

8. Zelner JL, Murray MB, Becerra MC, Galea J, Lecca L, Calderon R, et al. Bacillus Calmette–Guerin and isoniazid preventive therapy protect contacts of patients with tuberculosis. Am J Respir Crit Care Med. 2014;189(7):853–9.

9. Tuberculosis Research Centre, Indian Council of Medical Research. Risk of tuberculosis among contacts of isoniazid–resistant and isoniazid–susceptible cases. Int J Tuberc Lung Dis. 2011;15(6):782–8.

10. Radhakrishna S, Frieden TR, Subramani R, Santha T, Narayanan PR, Indian Council of Medical R. Additional risk of developing TB for household members with a TB case at home at intake: a 15–year study. Int J Tuberc Lung Dis. 2007;11(3):282–8.

11. Narain R, Nair SS, Rao GR, Chandrasekhar P. Distribution of tuberculous infection and disease among households in a rural community. Bull World Health Organ. 1966;34(4):639–54.

12. WHO Tuberculosis Chemotherapy Centre. An investigation of household contacts of open cases of pulmonary tuberculosis amongst the Kikuyu in Kiambu, Kenya. Bull World Health Organ. 1961;25(6):831–50.

13. Andrews RH, Devadatta S, Fox W, Radhakrishna S, Ramakrishnan CV, Velu S. Prevalence of tuberculosis among close family contacts of tuberculous patients in South India, and influence of segregation of the patient on early attack rate. Bull World Health Organ. 1960;23:463–510.

14. Loudon RG, Williamson J, Johnson JM. An analysis of 3,485 tuberculosis contacts in the city of Edinburgh during 1954–1955. Am Rev Tuberc. 1958;77(4):623–43.



15. Triasih R, Robertson C, Duke T, Graham SM. Risk of infection and disease with Mycobacterium tuberculosis among children identified through prospective community–based contact screening in Indonesia. Trop Med Int Health. 2015;20(6):737–43.

16. Whalen CC, Zalwango S, Chiunda A, Malone L, Eisenach K, Joloba M, et al. Secondary attack rate of tuberculosis in urban households in Kampala, Uganda. PLoS One. 2011;6(2):e16137.

17. Lewinsohn DA, Zalwango S, Stein CM, Mayanja–Kizza H, Okwera A, Boom WH, et al. Whole blood interferon–gamma responses to mycobacterium tuberculosis antigens in young household contacts of persons with tuberculosis in Uganda. PLoS One. 2008;3(10):e3407.

18. Amanullah F, Ashfaq M, Khowaja S, Parekh A, Salahuddin N, Lotia–Farrukh I, et al. High tuberculosis prevalence in children exposed at home to drug–resistant tuberculosis. Int J Tuberc Lung Dis. 2014;18(5):520–7.

19. Ma N, Zalwango S, Malone LL, Nsereko M, Wampande EM, Thiel BA, et al. Clinical and epidemiological characteristics of individuals resistant to M. tuberculosis infection in a longitudinal TB household contact study in Kampala, Uganda. BMC Infect Dis. 2014;14:352.

20. Rathi SK, Akhtar S, Rahbar MH, Azam SI. Prevalence and risk factors associated with tuberculin skin test positivity among household contacts of smear–positive pulnionary tuberculosis cases in Umerkot, Pakistan. Int J Tuberc Lung Dis. 2002;6(10):851–7.

21. Lienhardt C, Fielding K, Sillah J, Tunkara A, Donkor S, Manneh K, et al. Risk factors for tuberculosis infection in sub–Saharan Africa: a contact study in The Gambia. Am J Respir Crit Care Med. 2003;168(4):448–55.

22. Jones–Lopez EC, White LF, Kirenga B, Mumbowa F, Ssebidandi M, Moine S, et al. Cough Aerosol Cultures of Mycobacterium tuberculosis: Insights on TST / IGRA Discordance and Transmission Dynamics. PLoS One. 2015;10(9):e0138358.

23. Kifai EJ, Bakari M. Mantoux skin test reactivity among household contacts of HIV–infected and HIV un–infected patients with sputum smear positive TB in Dar es Salaam, Tanzania. East Afr J Public Health. 2009;6(2):211–8.

24. Nunn P, Mungai M, Nyamwaya J, Gicheha C, Brindle RJ, Dunn DT, et al. The effect of human immunodeficiency virus type–1 on the infectiousness of tuberculosis. Tuber Lung Dis. 1994;75(1):25–32.

25. Espinal MA, Perez EN, Baez J, Henriquez L, Fernandez K, Lopez M, et al. Infectiousness of Mycobacterium tuberculosis in HIV–1–infected patients with tuberculosis: a prospective study. Lancet. 2000;355(9200):275–80.

26. Hesseling AC, Mandalakas AM, Kirchner HL, Chegou NN, Marais BJ, Stanley K, et al. Highly discordant T cell responses in individuals with recent exposure to household tuberculosis. Thorax. 2009;64(10):840–6.

27. Guwatudde D, Nakakeeto M, Jones–Lopez EC, Maganda A, Chiunda A, Mugerwa RD, et al. Tuberculosis in household contacts of infectious cases in Kampala, Uganda. Am J Epidemiol. 2003;158(9):887–98.

28. Lees AW, Allan GW, Smith J, Tyrrell WF. Pulmonary tuberculosis in contacts: a ten year survey. Dis Chest. 1961;40:516–21.

29. Ahmad Khan F, Verkuijl S, Parrish A, Chikwava F, Ntumy R, El–Sadr W, et al. Performance of symptom–based tuberculosis screening among people living with HIV: not as great as hoped. AIDS. 2014;28(10):1463–72.

30. Nguyen DT, Bang ND, Hung NQ, Beasley RP, Hwang LY, Graviss EA. Yield of chest radiograph in tuberculosis screening for HIV–infected persons at a district–level HIV clinic. Int J Tuberc Lung Dis. 2016;20(2):211–7.

31. den Boon S, White NW, van Lill SW, Borgdorff MW, Verver S, Lombard CJ, et al. An evaluation of symptom and chest radiographic screening in tuberculosis prevalence surveys. Int J Tuberc Lung Dis. 2006;10(8):876–82.

63

32. Ministry of Health. Report on National TB Prevalence Survey 2009-2010. Nay Pyi Taw, Ministry of Health, Department of Health, 2012.

33. van’t Hoog AH, Meme HK, Laserson KF, Agaya JA, Muchiri BG, Githui WA, et al. Screening strategies for tuberculosis prevalence surveys: the value of chest radiography and symptoms. PLoS One. 2012;7(7):e38691.

34. Kapata N, Chanda–Kapata P, Ngosa W, Metitiri M, Klinkenberg E, Kalisvaart N, et al. The Prevalence of Tuberculosis in Zambia: Results from the First National TB Prevalence Survey, 2013–2014. PLoS One. 2016;11(1):e0146392.

35. Kebede AH, Alebachew Z, Tsegaye F, Lemma E, Abebe A, Agonafir M, et al. The first population–based national tuberculosis prevalence survey in Ethiopia, 2010–2011. Int J Tuberc Lung Dis. 2014;18(6):635–9.

36. Senkoro M, Mfinanga S, Egwaga S, Mtandu R, Kamara DV, Basra D, et al. Prevalence of pulmonary tuberculosis in adult population of Tanzania: a national survey, 2012. Int J Tuberc Lung Dis. 2016;20(8):1014–21.

37. Law I, Sylavanh P, Bounmala S, Nzabintwali F, Paboriboune P, Iem V, et al. The first national tuberculosis prevalence survey of Lao PDR (2010–2011). Trop Med Int Health. 2015;20(9):1146–54.

38. Adetifa IM, Kendall L, Bashorun A, Linda C, Omoleke S, Jeffries D, et al. A tuberculosis nationwide prevalence survey in Gambia, 2012. Bull World Health Organ. 2016;94(6):433–41.

39. Ayles H, Schaap A, Nota A, Sismanidis C, Tembwe R, De Haas P, et al. Prevalence of tuberculosis, HIV and respiratory symptoms in two Zambian communities: implications for tuberculosis control in the era of HIV. PLoS One. 2009;4(5):e5602.

40. Corbett EL, Zezai A, Cheung YB, Bandason T, Dauya E, Munyati SS, et al. Provider–initiated symptom screening for tuberculosis in Zimbabwe: diagnostic value and the effect of HIV status. Bull World Health Organ. 2010;88(1):13–21.

41. Datta M, Radhamani MP, Sadacharam K, Selvaraj R, Rao DL, Rao RS, et al. Survey for tuberculosis in a tribal population in North Arcot District. Int J Tuberc Lung Dis. 2001;5(3):240–9.

42. Gopi PG, Subramani R, Radhakrishna S, Kolappan C, Sadacharam K, Devi TS, et al. A baseline survey of the prevalence of tuberculosis in a community in south India at the commencement of a DOTS programme. Int J Tuberc Lung Dis. 2003;7(12):1154–62.

43. Ministry of Health. Report National TB Prevalence Survey, 2002 Cambodia. Phnom Penh, National Tuberculosis Control Programme, 2005.

44. Mathad JS, Bhosale R, Balasubramanian U, Kanade S, Mave V, Suryavanshi N, et al. Quantitative IFN–gamma and IL–2 Response Associated with Latent Tuberculosis Test Discordance in HIV–infected Pregnant Women. Am J Respir Crit Care Med. 2016;193(12):1421–8.

45. Rangaka MX, Wilkinson RJ, Boulle A, Glynn JR, Fielding K, van Cutsem G, et al. Isoniazid plus antiretroviral therapy to prevent tuberculosis: a randomised double–blind, placebo–controlled trial. Lancet. 2014;384(9944):682–90.



48. Sharma SK, Vashishtha R, Chauhan LS, Sreenivas V, Seth D. Comparison of TST and IGRA in Diagnosis of Latent Tuberculosis Infection in a High TB–Burden Setting. PLoS One. 2017;12(1):e0169539.



49. Spyridis NP, Spyridis PG, Gelesme A, Sypsa V, Valianatou M, Metsou F, et al. The effectiveness of a 9–month regimen of isoniazid alone versus 3– and 4–month regimens of isoniazid plus rifampin for treatment of latent tuberculosis infection in children: results of an 11–year randomized study. Clin Infect Dis. 2007;45(6):715–22.


51. van Zyl S, Marais BJ, Hesseling AC, Gie RP, Beyers N, Schaaf HS. Adherence to anti–tuberculosis chemoprophylaxis and treatment in children. Int J Tuberc Lung Dis. 2006;10(1):13–8.


53. Sterling TR, Scott NA, Miro JM, Calvet G, La Rosa A, Infante R, et al. Three months of weekly rifapentine and isoniazid for treatment of Mycobacterium tuberculosis infection in HIV–coinfected persons. AIDS. 2016;30(10):1607–15.

54. Sterling TR, Villarino ME, Borisov AS, Shang N, Gordin F, Bliven–Sizemore E, et al. Three Months of Rifapentine and Isoniazid for Latent Tuberculosis Infection. New England Journal of Medicine. 2011;365(23):2155–66.

55. Villarino ME, Scott NA, Weis SE, Weiner M, Conde MB, Jones B, et al. Treatment for preventing tuberculosis in children and adolescents: a randomized clinical trial of a 3–month, 12–dose regimen of a combination of rifapentine and isoniazid. JAMA Pediatr. 2015;169(3):247–55.

56. Bamrah S, Brostrom R, Dorina F, Setik L, Song R, Kawamura LM, et al. Treatment for LTBI in contacts of MDR–TB patients, Federated States of Micronesia, 2009–2012. Int J Tuberc Lung Dis. 2014;18(8):912–8.

57. Garcia–Prats AJ, Zimri K, Mramba Z, Schaaf HS, Hesseling AC. Children exposed to multidrug–resistant tuberculosis at a home–based day care centre: a contact investigation. Int J Tuberc Lung Dis. 2014;18(11):1292–8.

58. Schaaf HS, Garcia–Prats AJ, Hesseling AC, Seddon JA. Managing multidrug–resistant tuberculosis in children: review of recent developments. Curr Opin Infect Dis. 2014;27(3):211–9.


For further information, please contact:

World Health Organization20, Avenue Appia CH-1211 Geneva 27 SwitzerlandGlobal TB ProgrammeWeb site: www.who.int/tb

ISBN 978-92-4-155023-9

Latent tuberculosis



EXECUTIVE SUMMARY

1

Executive summary

Latent tuberculosis infection (LTBI) is defined as a state of persistent immune response to stimulation by Mycobacterium tuberculosis antigens with no evidence of clinically manifest active TB. There is no gold standard test for LTBI. The WHO guidelines on LTBI consider the probability of progression to active TB disease in a specific risk group, the epidemiology and burden of TB, the availability of resources and the likelihood of a broad public health impact. Two fragmented recommendations have been made for the management of LTBI, which resulted in a number of guideline documents, posing a challenge to implementation. Therefore, several WHO Member States requested consolidated guidelines on LTBI management.

The updated, consolidated guidelines in this document respond to that request. They provide a comprehensive set of WHO recommendations for programmatic management of LTBI and the basis and rationale for national guidelines. These guidelines supersede previous WHO policy documents on the management of LTBI in people living with HIV and household contacts of people with TB and other at-risk groups. The guidelines were prepared in accordance with the requirements and recommended process of the WHO Guideline Review Committee. Seven systematic reviews were conducted to update the recommendations and make new ones. The Guideline Development Group considered the quality of the evidence, benefits and harms, values and preferences, equity, costs, acceptability and feasibility of implementation in formulating the recommendations and determining their strength.

The recommendations are presented logically according to the cascade of care for managing LTBI: identification of at-risk populations (adults and children living with HIV, HIV-negative adult and child contacts and other HIV-negative at-risk groups), ruling out active TB disease, testing for LTBI, providing treatment, monitoring adverse events, adherence and completion of treatment and monitoring and evaluation. The recommendations are categorized as: existing ones previously approved by the review committee and published, which are still valid; updated recommendations that were previously approved by the review committee but for which the evidence was reviewed, discussed with the Guidelines Development Group (GDG) and updated (including for clarity); and new recommendations. There are 10 existing, 7 updated and 7 new recommendations.

In general, the GDG reviewed the evidence from the systematic reviews and discussed each population risk group identified in detail for the prevalence of LTBI, the risk for progression to active TB and the incidence of active TB as compared with that in the general population. The GDG used the guiding principle that individual benefit outweighs risk as the mainstay of recommendations on LTBI testing and treatment. The GDG found clear evidence for the benefit of systematic testing and treatment of LTBI for people living with HIV and infants and children under 5 years of age who are household contacts of pulmonary TB patients, in all settings, regardless of the background epidemiology of TB. Similarly, they concluded that HIV-negative groups at clinical risk, such as patients initiating anti-TNF treatment, receiving dialysis, preparing for organ or haematological transplantation and those with silicosis would also benefit from testing and treatment of LTBI, regardless of the background TB epidemiology, because of their increased risk of progression to active TB disease.

The specific recommendations are given below.

A. Identification of at-risk populations for LTBI testing and treatment Adults, adolescents, children and infants living with HIV

�� Adults and adolescents living with HIV, with unknown or a positive tuberculin skin test (TST) and are unlikely to have active TB should receive preventive treatment of TB as part of a comprehensive package of HIV care. Treatment should be given to these individuals irrespective of the degree of immunosuppression

EXECUTIVE SUMMARY


and also to those on antiretroviral treatment (ART), those who have previously been treated for TB and pregnant women. (Strong recommendation, high-quality evidence. Existing recommendation)




HIV-negative household contacts�� HIV-negative children aged < 5 years who are household contacts of people with bacteriologically confirmed

pulmonary TB and who are found not to have active TB on an appropriate clinical evaluation or according to national guidelines should be given TB preventive treatment. (Strong recommendation, high-quality evidence. Updated recommendation)



Other HIV-negative at-risk groups�� Patients initiating anti-TNF treatment, patients receiving dialysis, patients preparing for an organ or

haematological transplant and patients with silicosis should be systematically tested and treated for LTBI. (Strong recommendation, low–very low-quality evidence. Updated recommendation)



B. Algorithms to rule out active TB disease�� Adults and adolescents living with HIV should be screened for TB according to a clinical algorithm. Those

who do not report any of the symptoms of current cough, fever, weight loss or night sweats are unlikely to have active TB and should be offered preventive treatment, regardless of their ART status. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Chest radiography may be offered to people living with HIV and on ART and preventive treatment given to those with no abnormal radiographic findings. (Conditional recommendation, low-quality evidence. New recommendation)

�� Adults and adolescents living with HIV who are screened for TB according to a clinical algorithm and who report any of the symptoms of current cough, fever, weight loss or night sweats may have active TB

3

and should be evaluated for TB and other diseases that cause such symptoms. (Strong recommendation, moderate-quality evidence. Updated recommendation)

�� Infants and children living with HIV who have poor weight gain, fever or current cough or who have a history of contact with a case of TB should be evaluated for TB and other diseases that cause such symptoms. If the evaluation shows no TB, these children should be offered preventive treatment, regardless of their age. (Strong recommendation, low-quality evidence. Updated recommendation)

�� The absence of any symptoms of TB and the absence of abnormal chest radiographic findings may be used to rule out active TB disease among HIV-negative household contacts aged ≥ 5 years and other at-risk groups before preventive treatment. (Conditional recommendation, very low-quality evidence. New recommendation)

C. Testing for LTBI�� Either a tuberculin skin test (TST) or interferon-gamma release assay (IGRA) can be used to test for LTBI.

(Strong recommendation, very low-quality evidence. New recommendation)



D. Treatment options for LTBI�� Isoniazid monotherapy for 6 months is recommended for treatment of LTBI in both adults and children

in countries with high and low TB incidence. (Strong recommendation, high-quality evidence. Existing recommendation)




�� In settings with high TB incidence and transmission, adults and adolescents living with HIV who have an unknown or a positive TST and are unlikely to have active TB disease should receive at least 36 months of IPT, regardless of whether they are receiving ART. IPT should also be given irrespective of the degree of immunosuppression, history of previous TB treatment and pregnancy. (Conditional recommendation, low-quality evidence. Existing recommendation).

E. Preventive treatment for contacts of patients with multidrug-resistant-TB�� In selected high-risk household contacts of patients with multidrug-resistant tuberculosis, preventive

treatment may be considered based on individualised risk assessment and a sound clinical justification. (Conditional recommendation, very low-quality evidence. New recommendation)

EXECUTIVE SUMMARY


Important additional considerationsAdverse events monitoring The risk for adverse events during preventive treatment must be minimized. Individuals receiving treatment for LTBI should be monitored routinely and regularly at monthly visits to health care providers. The prescribing health care provider should explain the disease process and the rationale for the treatment and emphasize the importance of completing it. People receiving treatment should be urged to contact their health care providers if they develop symptoms between visits, such as anorexia, nausea, vomiting, abdominal discomfort, persistent fatigue or weakness, dark-coloured urine, pale stools or jaundice. If a health care provider cannot be consulted at the onset of such symptoms, the patient should immediately stop treatment.

Adherence and completion of preventive treatment Adherence to the full course and completion of treatment are important determinants of clinical benefit, both to the individual and to the success of the programme. Interventions should be tailored to the specific needs of the risk groups and to the local context to ensure adherence and completion of treatment.

Programmatic management, monitoring and evaluationThe national programme should prepare a national plan for programmatic management of LTBI, including prioritizing groups identified as being at high risk on the basis of local epidemiology and the health system. They should create a conducive environment for the policy and the programme, including national and local policies and standard operating procedures to facilitate implementation of the recommendations in these guidelines. Programmatic management of LTBI should include monitoring and evaluation systems that are aligned with national systems for patient monitoring and surveillance. Appropriate recording and reporting tools should be developed, with standardized indicators.



ISBN 978-92-4-155023-9

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENTA

Latent tuberculosis



ANNEX 3Values and preferences

for the management of latent tuberculosis infection:

survey of populations affected by the recommendations

Latent tuberculosis infection Updated and consolidated guidelines for

programmatic managementAnnex 3. Values and preferences for the

management of latent tuberculosis infection: survey of populations affected by the recommendations












WHO/CDS/TB/2018.9


BackgroundManagement of latent tuberculosis infection (LTBI) is critical to end the global tuberculosis (TB) epidemic, as stated in the WHO End TB strategy. WHO currently recommends two approaches for the management of LTBI, based on TB incidence and on income (1). The recent WHO guidelines recommend several LTBI treatment options for a wide range of risk groups in countries with a low burden of TB, defined as high- or upper–middle-income countries with an estimated TB incidence rate of < 100 per 100 000 population (1). For countries with a high TB burden, WHO recommends isoniazid as preventive therapy for people living with HIV and for children aged < 5 years who are household contacts of TB cases (2,3). At the request of Member States, WHO is preparing a unique, comprehensive document to guide policy for the management of LTBI in countries with high and low TB burdens.

Understanding the values and preferences of populations affected by the recommendations is essential for determining the strength of the recommendations (4). Values and preferences are particularly important in the management of LTBI, as treatment is given to asymptomatic individuals with no immediate harm due to TB infection in order to reduce their risk for active TB (5). We therefore conducted an online survey of the values and preferences for LTBI management of populations affected by the recommendations.

Methods We developed an online survey to solicit values and preferences for the management of LTBI on Survey Monkey (Palo Alto, CA, USA; www. surveymonkey.com). The survey contained two types of question: preferences for receiving LTBI treatment and features of the treatment. Background information was provided, including the concept of LTBI and the benefits and harm associated with its treatment. Three groups of respondents were asked about their preferences for LTBI treatment: people living with HIV, household contacts of a person with active TB disease and household contacts of a person with active MDR-TB disease. People with children were also asked about their preferences for LTBI treatment for their children.

The survey was in English and was disseminated through the WHO ENGAGE-TB listserve, which has members working on TB within nongovernmental and civil society organizations. It was also disseminated through networks of members of the Guideline Development Group who represent civil society organizations. Participation in the survey was voluntary and anonymous. The survey was open, and data were collected for 2 weeks in February 2017.

Data were exported into Excel spreadsheets, and descriptive analyses were performed with STATA, version 13.1.

Results Of the 174 participants, 142 completed at least one question about their preferences and were included in the analysis. Their characteristics are shown in Table 1. The median age was 46 years (interquartile range [IQR]: 37–54 years). Participants in the African Region accounted for 35.9% of respondents. Only 12 (7.0%) respondents reported that they were HIV-positive.

People with HIV

Of the 12 respondents with HIV, 8 reported they would strongly or somewhat prefer to receive preventive treatment (Table 2). One respondent stated he/she would “somewhat not prefer” because of “additional pill burden and toxicity with reduced efficacy”. All the features of preventive treatment were considered very important or important by 80–100% of the respondents (Table 3).

Household contacts of a person with active TB disease

More than 80% (n = 113) of the respondents said that they would strongly or somewhat prefer to receive TB preventive treatment if they had contact with a person with active TB in their household (Table 2). Similarly, of


59 respondents with children, more than 80% would strongly or somewhat prefer to give preventive treatment to their children, regardless of their age. The reasons for not preferring preventive treatment included: concern about side-effects and drug-resistant TB, insufficient efficacy, low risk for active TB, duration of treatment and belief that treatment is unnecessary for healthy individuals or those without active disease.

More than 90% of the respondents considered that the following features of preventive treatment are very important or important: shorter duration, fewer side-effects, fewer visits to a clinic and fewer pills (Table 4). Fewer respondents rated “less frequent intake” and “no need for directly observed therapy (DOT)” as very important or important (77.3% and 74.4%, respectively). Similarly, 90–100% of the respondents rated all the features as very important or important for their children, except for “no need for DOT” (Table 5).

Household contacts of a person with MDR-TB disease

More than 80% of respondents would strongly or somewhat prefer to receive preventive treatment or give it to their children if they were exposed to someone with MDR-TB disease in their household. The reasons for not preferring preventive treatment included: limited evidence of the efficacy of preventive treatment for MDR-TB and concern about side-effects and the development of drug resistance.

DiscussionThe survey consistently showed a strong preference for TB preventive treatment, even against MDR-TB, despite the limited evidence. The strong preferences should be taken into account in making recommendations. As expected, shorter duration and fewer side-effects were considered important attributes of preventive treatment, as they may result in shorter regimens, such as 3 months’ weekly rifapentine plus isoniazid, which also has fewer side-effects than isoniazid monotherapy. No requirement for DOT was considered another important attribute, to a similar extent as shorter duration and fewer side-effects. The 2015 WHO LTBI guidelines state that “The (Guideline) Panel noted that the 3-month weekly rifapentine plus isoniazid regimen should be given under direct observation as the evidence available so far is based on this circumstance” (1). A need for DOT could be a barrier to the scale-up of the weekly regimen.

The survey had several limitations. First, self-selection bias might have resulted in an overestimate of the preference for preventive treatment. It is possible that individuals who supported provision of preventive treatment were more likely to participate in the survey. Secondly, the results may be of limited generalizability, as the survey was conducted in English, and only people with access to the Internet could participate. Thirdly, a history of TB and treatment for TB might have affected the preference of respondents for preventive treatment, and we did not collect data on history of TB. Fourthly, there were few participants with HIV infection.

In conclusion, preventive treatment was strongly preferred by the participants in the survey. This should be taken into account in determining the strength of the recommendations.

References1. Guidelines on the management of latent tuberculosis infection. Geneva: World Health Organization; 2015.2. Guidelines for intensified tuberculosis case-finding and isoniazid preventive therapy for people living with

HIV in resource-constrained settings. Geneva: World Health Organization; 2011.3. Recommendations for investigating contacts of persons with infectious tuberculosis in low- and middle-

income countries.Geneva: World Health Organization; 2012.4. WHO handbook for guideline development. 2nd edition. Geneva: World Health Organization; 2014.5. Denholm JT, Matteelli A, Reis A. Latent tuberculous infection: ethical considerations in formulating public

health policy. Int J Tuberc Lung Dis. 2015;19:137–40.6. Belknap R, Borisov A, Holland D, Feng PJ, Millet JP, Martinson N et al. Adherence to once-weekly self-

administered isoniazid and rifapentine for latent TB infection: iAdhere. In: Conference on Retroviruses and Opportunistic Infections; Boston (MA); 2016:Abstract 827LB (ww.croiconference.org/sessions/adherence-once-weekly-self-administered-inh-and-rifapentine-latent-tb-iadhere, accessed 27 December 2017).


Table 1. Characteristics of respondents (n = 142)

Age (years), median (IQR) 46 (37–54)

Gender, n (%)

Female 64 (45.1)

Male 77 (54.2)

Prefer not to indicate 1 (0.7)

WHO region, n (%)

African 51 (35.9)

Americas 25 (17.6)

Eastern Mediterranean 7 (4.9)

European 28 (19.7)

South-East Asian 26 (18.3)

Western Pacific 5 (3.5)

HIV status, n (%)

Positive 10 (7.0)

Negative 127 (89.4)

Prefer not to indicate 5 (3.5)

Living with children,a n (%)

Yes 62 (44.6)

No 77 (55.4)

IQR, interquartile rangea Data were missing for three participants.

Table 2. Preferences for TB preventive treatment

Strongly prefer, n (%)

Somewhat prefer, n (%)

Neither prefer nor not prefer,

n (%)

Somewhat not prefer, n (%)

Strongly not prefer, n (%)

People living with HIV

Would prefer to receive (n=10) 7 (70.0) 1 (10.0) 1 (10.0) 1 (10.0) 0 (0.0)

Household contact of a person with active TB disease


Would prefer to give to your children

0–4 years (n=59) 38 (64.4) 11 (18.6) 1 (1.7) 5 (8.5) 4 (6.8)

5–9 years (n=59) 33 (55.9) 15 (25.4) 2 (3.4) 6 (10.2) 3 (5.1)

10–14 years (n=59) 31 (52.5) 16 (27.1) 2 (3.4) 6 (10.2) 4 (6.8)

Household contact of a person with MDR-TB disease


Would prefer to give to your children

0–4 years (n=59) 31 (57.4) 13 (24.1) 3 (5.6) 3 (5.6) 4 (7.4)

5–9 years (n=59) 30 (55.6) 14 (25.9) 3 (5.6) 4 (7.4) 3 (5.6)

10–14 years (n=59) 33 (61.1) 12 (22.2) 2 (3.7) 3 (5.6) 4 (7.4)


Table 3. Important features of preventive treatment for people living with HIV

Median scorea (IQR)

Very important, n

(%)

Important, n (%)

Somewhat important, n

(%)

Less important, n

(%)

Not important, n

(%)

Shorter duration 5 (4–5) 6 (60.0) 4 (40.0) 0 0 0

Less frequent intake 4.5 (4–5) 5 (50.0) 3 (30.0) 0 0 2 (20.0)

Fewer side-effects 5 (4–5) 7 (70.0) 3 (30.0) 0 0 0

Fewer visits to the clinic 5 (5–5) 8 (80.0) 1 (10.0) 1 (10.0) 0 0

Fewer pills 5 (4–5) 5 (50.0) 4 (40.0) 1 (10.0) 0 0

No need for DOT 5 (4–5) 7 (70.0) 3 (30.0) 0 0 0

No need to change dosage of anti-HIV drugs 4.5 (4–5) 5 (50.0) 3 (30.0) 1 (10.0) 0 1 (10.0)

HIV: human immunodeficiency virus; IQR: interquartile range; DOT: directly observed therapya 5-point rating scale where 5 is “very important” and 1 is “not important”.

Table 4. Important features of preventive treatment for adult household contacts

Median scorea (IQR)

Very important, n

(%)

Important, n (%)


(%)

Less important, n

(%)

Not important, n

(%)

Shorter duration 5 (5–5) 116 (82.3) 23 (16.3) 2 (1.4) 0 0

Less frequent intake 4 (4–5) 68 (48.2) 41 (29.1) 14 (9.9) 12 (8.5) 6 (4.3)

Fewer side-effects 5 (5–5) 119 (84.4) 20 (14.2) 2 (1.4) 0 0

Fewer visits to the clinic 5 (4–5) 89 (63.1) 44 (31.2) 5 (3.6) 2 (1.4) 1 (0.7)

Fewer pills 5 (4–5) 90 (63.8) 38 (27.0) 10 (7.1) 3 (2.1) 0

No need for DOT 5 (3–5) 85 (56.7) 25 (17.7) 23 (16.3) 4 (2.8) 9 (6.4)

IQR: interquartile range; DOT: directly observed therapya 5-point rating scale where 5 is “very important” and 1 is “not important”.

Table 5. Important features of preventive treatment for child household contacts

Median scorea (IQR)

Very important, n

(%)

Important, n (%)


(%)

Less important, n

(%)

Not important, n

(%)

Shorter duration 5 (5–5) 50 (84.8) 9 (15.3) 0 0 0

Less frequent intake 5 (4–5) 41 (69.5) 14 (23.7) 3 (5.1) 0 1 (1.7)

Fewer side-effects 5 (5–5) 54 (91.5) 5 (8.5) 0 0 0

Fewer visits to the clinic 5 (4–5) 41 (69.5) 14 (23.7) 2 (3.4) 2 (3.4) 0

Easy to swallow (e.g. water-dispersible) 5 (4–5) 52 (88.1) 7 (11.9 ) 0 0 0

No need for DOT 5 (4–5) 37 (62.7) 9 (15.3) 6 (10.2) 3 (5.1) 4 (6.8)

IQR: interquartile range; DOT: directly observed therapya 5-point rating scale where 5 is “very important” and 1 is “not important”.



Latent tuberculosis



ANNEX 2Evidence-to-Decision and

GRADE tables

Latent tuberculosis infection Updated and consolidated guidelines for programmatic management

Annex 2. Evidence-to-Decision and GRADE tables












WHO/CDS/TB/2018.8

LATENT TUBERCULOSIS INFECTION: UPDATED AND CONSOLIDATED GUIDELINES FOR PROGRAMMATIC MANAGEMENT: ANNEX 2. EVIDENCE-TO-DECISION AND GRADE TABLES 3

PICO1: What is the prevalence of LTBI, risk of progression to active TB and cumulative prevalence of active TB among household contacts without HIV in different age groups in high TB incidence countries?

Problem Identification of household contacts for diagnosis and treatment of LTBI

Background For programmatic LTBI management, the risk associated with diagnosing and treating LTBI should be weighed against the benefit. Mass population screening and treatment of LTBI are not feasible, because of insensitive tests, high cost, poor sustainability, uncertain cost–effectiveness and risks for serious and fatal side-effects. Therefore, populations at high risk for active TB should be targeted. Accordingly, WHO currently recommends systematic LTBI screening and treatment for children < 5 years who are household contacts of TB cases in high-TB incidence countries with limited resources. Systematic LTBI screening and treatment are also recommended for children aged ≥ 5 years, adolescents and adults in low-TB incidence countries.Three systematic reviews were undertaken to determine whether the target age group should be extended in high-TB incidence countries by measuring three outcomes among household contacts in different age groups: prevalence of LTBI, risk of progression to active TB and cumulative prevalence of active TB. These outcomes were selected because the risk for TB may reflect a higher prevalence of LTBI and an increased risk for progression from LTBI to active TB.

Option Systematic screening and treatment for LTBI among household contacts in specific age groups

Comparison NA

Main outcomes Prevalence of LTBI, risk of progression to active TB and cumulative prevalence of active TB among household contacts in different age groups

Setting High-TB incidence countries (estimated TB incidence rate ≥ 100 per 100 000)

Perspective Health system and public health

AssessmentJudgement Research evidence Additional considerations

Prob

lem

Is the problem a priority? No Yes Varies Don’t know

Globally in 2015, there were an estimated 10.4 million incident cases of TB and 1.8 million deaths from TB. Management of LTBI is critical in order to end the global TB epidemic, as stated in the WHO End TB Strategy. Active TB must be excluded before TB preventive treatment is given. Although WHO currently recommends systematic LTBI screening and treatment for household contacts of any age in low-TB incidence countries, it is recommended only for child household contacts < 5 years old in high-TB incidence countries.

Bala

nce

of e

ffect

s

Do the benefits outweigh the harms? Yes No They are equal Uncertain

We updated three systematic reviews conducted for the previous LTBI guidelines, focusing on household contacts. The first review addressed the prevalence of LTBI among household contacts by age group, the second the risk of progression from LTBI to active TB among household contacts and the third the cumulative prevalence of active TB among household contacts, irrespective of baseline LTBI status. In most of the studies, prevalent TB cases were those identified at the baseline visit, and those identified later were counted as incident cases. The incidence of TB therefore depended on the timing of the baseline visit relative to the diagnosis of the index case; focusing on incident TB cases, therefore, may introduce bias. In the second and the third reviews, both prevalent TB during the baseline visit and incident TB during follow-up were included in the numerator. We estimated the prevalence ratios by comparing the prevalence of LTBI among household contacts by age stratum, with children < 5 years as the reference group.


Bala

nce

of e

ffect

sPooled estimates of prevalence of LTBI among household contacts by age stratum as compared with children < 5 years in high-TB incidence countries (estimated TB incidence rate ≥ 100 per 100 000)

Age group (years) No. of studies (no. of participants) Prevalence ratio (95% CI)

0–4 – 1.0 (reference)

5–9 14 1.62 (1.25;2.11)

10–14 11 (18 033) 2.33 (1.55;3.5)

5–14 16 (13 867) 1.32 (1.11;1.56)

≥ 15 19 (28 725) 2.04 (1.53;2.63)

The analysis suggested that the prevalence of LTBI increases with age. Furthermore, we estimated risk ratios for:• development of active TB among household contacts with LTBI and • cumulative prevalence of active TB irrespective of baseline LTBI status, by age stratum, with children aged < 5 years as

the reference. The cumulative prevalence of active TB includes cases diagnosed during contact investigations at baseline and incident cases that developed thereafter. The table below summarizes the results of the two analyses.

Pooled estimates of risk for active TB among household contacts stratified by age and baseline LTBI status

Baseline LTBI status positive Regardless of baseline LTBI status

Age (years) No. of studies Risk ratio (95% CI) No. of studies Risk ratio (95% CI) (no. of participants) (no. of participants)

0–4 – 1.0 (reference) – 1.0 (reference)

5–14 4 (1959) 0.28 (0.12;0.65) 6 (7292) 0.39 (0.18;0.85)

≥15 3 (5 341) 0.22 (0.08;0.60) 4 (13 620) 0.68 (0.56;0.83)

The review consistently showed that older household contacts have lower risk of the development of active TB compared to children < 5 years. Furthermore, in the second and the third review, we compared the risk of active TB among household contacts stratified by age groups compared to the general population using year– adjusted national estimated TB incidence from the WHO.


Bala

nce

of e

ffect

sPooled estimates of risk of development of active TB among household contacts stratified by age and baseline LTBI status compared to the general population.

Baseline LTBI status positive Regardless of baseline LTBI status

Follow-up <12 months Follow-up <24 months Follow-up <12 months Follow-up <24 months

Age No. of studies Risk ratio #studies Risk ratio #studies Risk ratio #studies Risk ratio (years) (no. of participants) (no. of participants) (no. of participants) (no. of participants)

General population - 1.0 (Reference) - 1.0 (Reference) - 1.0 (Reference) - 1.0 (Reference)

0-4 2 (265) 24.32 (0.73-811.02) 3 (585) 22.87 (7.65-68.63) 3 (1 930) 25.86 (16.87-39.66) 5 (2 773) 14.8 (9.82-22.3)

5-9 1 (298) 30.98 (14.26-67.31) 1 (298) 15.49 (7.89-30.4) 1 (1 464) 18.39 (9.75-34.68) 1 (1 464) 9.2 (5.55-15.23)

10-14 1 (363) 55.1 (28.55-106.33) 1 (363) 27.55 (16.16-46.96) 1 (1 340) 25.83 (13.97-47.76) 1 (1 340) 12.92 (8.0-20.86)

5-14 2 (728) 27.13 (17.47-54.07) 3 (1 203) 8.22 (2.3-29.36) 3 (3 067) 24.11 (16.89-34.43) 5 (4 197) 6.29 (2.88-13.72)

≥15 1 (3 879) 30.74 (17.46-54.07) 2 (4 450) 13.35 (9.46-18.83) 1 (9 380) 24.68 (14.18-42.98) 3 (10 531) 11.67 (7.55-18.02)

The results showed that household contacts have substantially higher risk of active TB compared to the general population regardless of their age.

Cer

tain

ty o

f evi

denc

e What is the overall certainty of the evidence of effects? Very low Low Moderate High No included studies

Val

ues

Is there important uncertainty about or variability in how much people value the main outcomes? Important uncertainty

or variability No important

uncertainty or variability

Minimal uncertainty

We conducted an online survey (Annex 3) to solicit the values and preferences of individuals affected by the recommendations. Responses were available from 142 respondents with a median age of 46 years (IQR: 37–54 years). More than 80% of the respondents reported that they would strongly or somewhat prefer to receive TB preventive treatment if they were in contact with a person with active TB in the household. Similarly, of 59 respondents with children, more than 80% would strongly or somewhat prefer to give preventive treatment to their children, regardless of the children’s age.

Concern about whether the respondents in the online survey correctly reflect the values of clients.


Reso

urce

s re

quire

dHow large are the resource requirements (costs)? Greater resource

requirements with the intervention

Less resource requirements with the intervention

Neither greater nor less

Varies Don’t know

National programmes could build upon existing programmes for children < 5 years, which could reduce the additional resource requirements.

Cos

t effe

ctiv

enes

s

Does the cost–effectiveness of the intervention favour the intervention or the comparison? Favours the

comparison Favours neither the

intervention nor the comparison

Favours the intervention

Varies No included studies

A systematic review of the cost–effectiveness of management of LTBI was undertaken for the 2015 WHO LTBI guidelines. The review covered six studies on contacts of patients with active TB, all in low-TB incidence countries; none provide the specific age groups of contacts. These studies suggested that screening and treatment of LTBI among contacts may save costs for the health care system and/or have a favourable incremental cost–effectiveness ratio.

Cost–effectiveness data from low-TB incidence countries may not be applicable to high-TB incidence countries, where the risk for re-infection is high. However, the GDG noted data suggesting the durability of protection in high-TB incidence countries. A recent modelling study suggested that preventive treatment without LTBI testing is cost–effective for child contacts < 5 years old (1).

Equi

ty

What would be the impact on health equity? Reduced Increased Varies Don’t know


Acc

epta

bilit

yIs the intervention acceptable to key stakeholders? No Yes Varies Don’t know

Might be acceptable to key stakeholders, including health workers and programme managers; however, extension of the target age group might add a burden for national programmes that are struggling even to provide preventive treatment for child household contacts < 5 years.

Feas

ibili

ty

Is the intervention feasible to implement? No Yes Varies Don’t know

Depends on setting, health infrastructure (e.g. availability of test and drugs) and population groups (e.g. adolescents).

Summary of judgementsJudgement Implications

Problem No Yes Varies Unknown

Balance of effects No Equal Yes Uncertain

Certainty of evidence Very low Low Moderate High No included

studies

ValuesImportant


Minimal uncertainty

No important uncertainty or

variability

Resources required Greater Neither greater nor less Less Varies Unknown

Cost–effectiveness Favours the comparison

Favours neither the intervention

nor the comparison

Favours the intervention Varies No included

studies

Equity Reduced Increased Varies Unknown

Acceptability No Yes Varies Unknown

Feasibility No Yes Varies Unknown


ConclusionsWhat is the prevalence of LTBI, risk of progression to active TB, and cumulative prevalence of active TB among household contacts without HIV in different age groups in high TB incidence countries?

Recommendation In favour of☒

Against☐

No recommendation☐

Strength of recommendation

Strong☐

Conditional☒

Recommendation In countries with a high TB incidence, children aged ≥ 5 years, adolescents and adults who are household contacts of people with bacteriologically confirmed pulmonary TB who are found not to have active TB by an appropriate clinical evaluation or according to national guidelines may be given TB preventive treatment. (Conditional recommendation, low-quality evidence.) Remark: Appropriate clinical evaluation should include assessment of the intensity of and risk for exposure, the risk for development of active TB and/or ascertainment of infection by testing for LTBI.

Justification The GDG agreed that, overall, the potential benefits of preventive treatment for household contacts outweigh the harm, regardless of age, given the high risk for development of active TB disease. The GDG also noted that the balance of benefits and harm depends on confirmation of infection by LTBI testing, and the benefits would be greater in household contacts with a positive LTBI test.There was overall consensus that more resources would be required and lack of evidence on cost–effectiveness. A systematic review suggested that screening and treatment of LTBI among contacts may save costs for the health care system or have a favourable incremental cost–effectiveness ratio. However, six of the studies were conducted in low-TB incidence countries, and the GDG noted that the results are not applicable in high-TB incidence countries, where the risk for re-infection is high. The GDG also noted evidence for the durability of protection in high-TB incidence countries. The GDG further noted that national programmes could build upon existing programmes for children < 5 years, which could reduce the additional resources required. There was general consensus that preventive treatment for household contacts could be acceptable to key stakeholders, including health workers and programme managers, although extension of the target age group could add a burden to national programmes that are struggling even to implement preventive treatment for children < 5 years.

Subgroup considerations

Implementation considerations

In order to ensure that the benefits of preventive treatment outweigh the harm, careful clinical assessment of the intensity of and risk for exposure, of the risk for development of active TB and/or with LTBI testing are required. Active TB must be excluded before preventive treatment is given.It is important to provide support for adherence adapted to the local context to ensure completion of treatment. This may be particularly challenging for certain populations such as adolescents. The support should take into account their needs.National programmes should ensure the availability of tests and drugs and properly train health care workers to provide preventive treatment for household contacts of all ages.

Monitoring and evaluation

Research priorities Methods to improve adherence and completion rate. Implementation research to improve effectiveness and efficiency of managing household contacts (e.g. household-based intervention to reduce barriers). Development of diagnostic tests with improved performance and predictive value for reactivation of TB. Durability of protection by preventive treatment in TB endemic settings.


GRADE tables: SR1SR1. Risk for LTBI among household contacts by age stratum: high-TB incidence countries

Quality assessment No. LTBI+/no. tested EffectQuality Importance

No. of studies Design Risk of bias Inconsistency Indirectness Imprecision Comparator 0–5 years RR(95% CI)

Absolute per 1000 (95% CI)

AGE GROUPS COMPARED: 5–10 YEARS VS 0–5 YEARS

14 studies (2–15)

Cross-sectional Not serious1,2 Serious3 Not serious Not serious4 2265/

85071298/9526

1.62(1.25;2.11)

85.1(34.2;151.1) Moderate Important


11 studies (2,4,6,8,9,

10-15)

Cross-sectional Not serious5 Serious6 Not serious Not serious7 2616/

67821093/9005

2.33(1.55;3.5)



16 studies8 Cross-sectional Serious9 Serious10 Not serious Not serious11 3709/

87721605/5095

1.32(1.11;1.56)

99.7(34.9;176.5) Low Important

AGE GROUPS COMPARED: > 15 YEARS VS 0–5 YEARS

19 studies12 Cross-sectional Not serious13 Serious14 Not serious Not serious15 13218/

219621979/6763

2.04(1.53;2.63)


1 Potential selection bias in (3), as only 69% of participants were household contacts.2 Potential misclassification: Eight studies (4–6,8,11,12,14,15) did not indicate whether household contacts with active TB were excluded from the analysis or did not provide sufficient data for calculation of the number of household

contacts with active TB per age stratum. 3 High heterogeneity among studies (I2 = 94%), probably due to differences in background TB incidence. The risk ratios of two studies (2,6) showed opposite effects.4 Small sample size in (6) (n < 50).5 Potential misclassification: Reports of seven studies (4,6,8,11,12,14,15) did not indicate whether household contacts with active TB were excluded from the analysis or did not provide sufficient data for calculation of the number of

household contacts with active TB per age stratum.6 High heterogeneity among studies (I2 = 97%) probably due to differences in background TB incidence. The risk ratio in one study (6) showed opposite effect.7 Wide 95% CI of pooled risk ratio. Small sample size in (6) (n < 50) and (13) (n < 100).8 Studies included: (4,6,9,11,13,16–26).9 Potential selection bias in (17), as only 89% of participants were household contacts.10 High heterogeneity among studies (I2 = 93%), probably due to differences in background TB incidence. The risk ratios in three studies (6,19,21) showed opposite effects.11 Small sample size in (6) and (18) (n < 50).12 Studies included: (4–6,9–11,13–16,19–27).13 Potential misclassification: The reports of ten studies (4–6,11,14,15,20,21,24,27) did not indicate whether household contacts with active TB were excluded from the analysis or did not provide sufficient data for calculation of the

number of household contacts with active TB per age stratum.14 High heterogeneity among studies (I2 = 98%), probably due to differences in background TB incidence. 15 Small sample size in 6 and 27 (n < 100).


SR2SR2. Development of active TB disease in household contacts with LTBI in high-TB incidence countries

Quality assessment No. of contacts(active TB/LTBI) Effect

Quality ImportanceNo. of studies Design Risk of bias Limitations Inconsistency Indirectness Imprecision Comparator 0–5 years RR

(95% CI)Absolute per

1000 (95% CI)


4 (9,14,17,23) Cohort Not serious Not serious Serious1 Not serious Serious2 54/1329 73/630 0.28(0.12;0.65)

83.8 (40.3;102.3) Low Critical


3 (9,14,23) Cohort Not serious Not serious Serious3 Not serious Not serious 186/4746 73/595 0.22 (0.08;0.60)

95.5 (49.1;112.6) Moderate Critical

Because there were few studies in the other categories, only data from studies in high-TB incidence countries with a follow-up of 1–2 years are presented in the table. 1 Serious inconsistencies due to heterogeneity (I2 = 71%). One study showed an increased risk in the age group 5–15 years. This was not observed in the other studies. 2 Few events.3 High heterogeneity among studies (I2 = 89.3%), probably due to differences in background TB incidence and methods used for diagnosis of active TB.

SR3SR3. Cumulative prevalence of active TB in household contacts, irrespective of baseline LTBI status, in high-TB incidence countries

Quality assessmentNo. of contacts

(active TB/total no. of contacts)

Effect

Quality Importance

No. of studies Design Risk of bias Limitations Inconsistency Indirectness Imprecision Comparator 0–5 years RR (95% CI)

Absolute per 1000 (95%

CI)


6 (9,14,17,18, 23,28)1 Cohort Not serious Not serious Serious2 Not serious Not serious 131/4389 203/2903 0.39

(0.18;0.85)42.9

(10.6;57.6) Moderate Important


4 (9,14,23,28) Cohort Not serious Not serious Not serious Not serious Not serious 417/10856 192/2764 0.68

(0.56;0.83)22

(12.1;30.3) High Important

Because there were few studies in the other categories, only data from studies in high-TB incidence countries with a follow-up of 1–2 years are presented in the table. 1 One outlier study (29) was excluded because of uncertainty about the cases that were included (co-prevalent vs incident cases). 2 High heterogeneity among studies (I2 = 87.6%), probably due to differences in background TB incidence.


Comparison with the general population for SR2Development of active TB disease in household contacts with LTBI in high-TB incidence countriesComparison with the general population (follow-up, 12 months)

Quality assessment No. of contacts(active TB/no. LTBI) Effect

Quality ImportanceNo. of studies Design Risk of bias Inconsistency Indirectness Imprecision Comparator General

population1 RR


1000 (95% CI)

COMPARISON: HOUSEHOLD CONTACTS AGED 0–5 YEARS VS GENERAL POPULATION

2 (9,17) Cohort Serious2 Serious3 Not serious Very serious40/35 41/10 000 24.32

(0.73;811.02)63

(–0.7;2187.1) Very low Critical32/230 13/10 000


1 (9) Cohort Serious2 Not serious Not serious Serious6 12/298 13/10 000 30.98 (14.26;67.31)

39 (17.2;86.2) Low Critical


1 (9) Cohort Serious2 Not serious Not serious Serious6 26/363 13/10 000 55.1 (28.55; 106.33)

70.3 (35.8;136.9) Low Critical


2 (9,17) Cohort Serious2 Not serious5 Not serious Serious64/67 41/10 000 27.13

(17.47;54.07)70.5

(21.3;220.7) Low Critical38/661 13/10 000

COMPARISON: HOUSEHOLD CONTACTS AGED > 15 YEARS VS GENERAL POPULATION


38.7 (21.4;69) Low Critical

1 LTBI does not apply to the general population.2 Ascertainment bias highly likely. TB cases in the general population detected passively, while TB cases in the contacts detected actively; therefore, relative and absolute risks might be overestimated. The composition of the

general and the study populations differs (general population of all ages versus a specific age group). 3 High heterogeneity (I2 = 83.9%) among studies, probably due to differences in background TB incidence. 4 Serious imprecision with a wide 95% CI for the effect estimates, probably due to the small study size and number of outcome events. 5 I2 = 72.5%, indicating moderate heterogeneity, probably due to differences in background TB prevalence; however, there is a trend across age groups and studies. 6 Few events and wide 95% CI.


Development of active TB disease in household contacts with LTBI in high-TB incidence countriesComparison with the general population (follow-up ≤ 24 months)1

Quality assessment No. of contacts(Active TB/no. LTBI) Effect


population2RR


1000 (95% CI)


3 (9,17,23) Cohort Serious3 Serious4 Not serious Serious5

0/35 82/10 00022.87 (7.65;

68.63)108.6

(33;334.6) Very low Important26/320 41/10 000

32/230 26/10 000



37.7 (17.9;76.4) Low Important



69 (39.4;119.5) Low Important


3 (9,17,23) Cohort Serious3 Serious6 Not serious Serious5

4/67 82/10 0008.22

(2.3;29.36)35.8

(6.5;140.8) Very low Important6/475 41/10 000

38/661 26/10 000


2 (9,23) Cohort Serious3 Not serious7 Not serious Not serious26/571 41/10 000 13.35

(9.46;18.83)41.4

(28.3;59.7) Moderate Important155/3879 26/10 000

1 These comparisons are based on studies with a maximum follow-up of 24 months. The TB incidence in the general population was multiplied by a factor of 2 to estimate the number of cases occurring during 24 months. 2 LTBI does not apply to the general population. 3 Ascertainment bias highly likely, because TB cases in the general population detected passively, while TB cases in the contacts detected actively. As a result, the relative and absolute risks might be overestimated. The composition

of the general and study populations differs (general population of all ages versus a specific age group). The TB incidence in the population was estimated by multiplying the yearly notification rate by a factor of 2. 4 High heterogeneity among studies (I2 = 84.4%), probably due to differences in background TB incidence. 5 Few events and wide 95% CI.6 I2 = 88.1%, indicating high heterogeneity, probably due to differences in background TB prevalence; however, there is a trend across age groups and studies. 7 I2 = 16%.


Comparison with the general population for SR3Cumulative prevalence of active TB in household contacts, irrespective of baseline LTBI status, in high-TB incidence countriesComparison with the general population (follow-up of 12 months)

Quality assessment No. of contacts(active TB/total no. contacts) Effect


populationRR

(95% CI)Absolute risk per 1000 (95% CI)


3 (9,17,18) Cohort Serious1 Not serious2 Not serious Serious3

2/31 28/10 00025.86

(16.87;39.66) 68 (43.4;105.7) Low Important9/108 41/10 000

73/1791 13/10 000








3 (9,17,18) Cohort Serious1 Not serious2 Not serious Serious3

8/102 28/10 00024.11

(16.89;34.43)63.2

(43.4;91.4) Low Important16/161 41/10 000

80/2804 13/10 000


1 (9) Cohort Serious1 Not serious Not serious Not serious 301/9380 13/10 000 24.68 (14.18;42.98) 30.8 (17.1;54.6) Moderate Important

1 Ascertainment bias highly likely, because TB cases in the general population detected passively, while TB cases in the contacts detected actively. As a result, the relative and absolute risks might be overestimated. The composition

of the general and study populations differs (general population of all ages versus a specific age group). 2 I2 = 0%. 3 Few events and wide 95% CI.


Cumulative prevalence of active TB in household contacts, irrespective of baseline LTBI status, in high-TB incidence countriesComparison with the general population (follow-up of 24 months)1

Quality assessment No. of contacts(active TB/total no. contacts) Effect


populationRR

(95% CI)Absolute risk per 1000 (95% CI)


5 (9,17,18, 23,28) Cohort Serious2 Not serious3 Not serious Serious4

2/31 55/10 000

14.8 (9.82;22.3)

83.9 (53.6;129.5) Low Important

37/335 100/10 000

9/108 82/10 000

55/508 41/10 000

73/1791 26/10 000


1 (9) Cohort Serious2 Not serious Not serious Serious4 35/1464 26/10 000 9.2 (5.55;15.23) 21.3 (11.8;37) Low Important


1 (9) Cohort Serious2 Not serious Not serious Serious4 45/1340 26/10 000 12.92 (8.0;20.86) 31 (18.2;51.6) Low Important


5 (9,17,18, 23,28) Cohort Serious2 Serious5 Not serious Not serious

8/102 55/10 000

6.29 (2.88;13.72) 32.2 (11.4;77.4) Low Important

5/439 100/10 000

16/161 82/10 000

10/691 41/10 000

80/2804 26/10 000


3 (9,23,28) Cohort Serious2 Not serious6 Not serious Not serious

34/432 100/10 00011.67

(7.55;18.02) 59.4 (36.5;94.7) Moderate Important49/719 41/10 000

301/9380 26/10 000

1 These comparisons were made in studies with a maximum follow-up of 24 months. The TB incidence in the general population was multiplied by a factor of 2 to estimate the number of cases occurring during 24 months. 2 Ascertainment bias highly likely, because TB cases in the general population detected passively, while TB cases in the contacts detected actively. As a result, the relative and absolute risks might be overestimated. The composition

of the general and study populations differs (general population of all ages versus a specific age group), and the TB incidence in the population was estimated by multiplying the yearly notification rate by a factor of 2. 3 Moderate heterogeneity among studies (I2 = 67.1%), probably due to differences in background TB incidence. 4 Few events and wide 95% CI. 5 High heterogeneity among studies (I2 = 87.5%), probably due to differences in background TB incidence. 6 Moderate heterogeneity among studies (I2 = 72.5%), probably due to differences in background TB incidence.


PICO2: What is the accuracy of WHO symptomatic screening to exclude active TB in individuals with HIV on antiretroviral treatment (ART)?

Population: People living with HIV on ART Background Active TB must be excluded before TB preventive treatment is given. Since 2011, WHO has recommended use of a four-symptom screening rule – current cough, weight loss, night sweats and fever – to exclude active TB in people living with HIV before initiating TB preventive treatment. This policy has contributed to wider use of preventive treatment globally, with almost 1 million recipients in 2015. Since the recommendation was established in 2011, there has been a significant increase in coverage with ART, and recent studies have shown an additive effect of TB preventive treatment and ART.

Intervention: WHO-recommended four-symptom screening plus abnormal chest radiography.Positive symptom screening defined as presence of any of four symptoms; for adults and adolescents: cough of any duration, weight loss, night sweats or fever; for children: poor weight gain, fever, current cough or history of contact with a TB case.

Role of the test: Rule out active TB before giving preventive treatment.

Linked treatments: Screening negative ➞ TB preventive treatment.

Anticipated outcomes:

True positive: Correct identification of an individual with active TB who should have further investigations.False negative: Incorrect identification of an individual with active TB as not having TB.True negative: Correct identification of an individual as not having active TB.False positive: Incorrect identification of an individual as requiring further investigations when they are actually TB negative.

Setting: High-TB incidence countries (estimated TB incidence rate ≥ 100 per 100 000).

Perspective: Health system and public health.

Subgroups:



Prob

lem


TB is the most frequent cause of HIV/AIDS-related deaths worldwide, despite progress in access to ART. TB caused 0.4 million deaths among people living with HIV in 2015, representing one third of all HIV-related mortality. TB preventive treatment is one of the key collaborative activities against TB and HIV. Preventive treatment can reduce TB incidence by about 30% and by up to 60% among those with a positive TST. Active TB must be excluded before TB preventive treatment is given.

Test

acc

urac

y

How accurate is the test? Very inaccurate Inaccurate Accurate Very accurate Varies Don’t know

We conducted a systematic review to assess the performance of the WHO-recommended four-symptom screening rule to exclude active TB before preventive treatment in HIV-positive people. Where possible, subgroup analyses were conducted by ART status, as the aim of this review was to study the effect with ART.

Two studies provided data on the combination of chest radiography and the four-symptom screening rule in PLHIV on ART. Any chest radiography abnormality was used in one study and chest radiography abnormality suggestive of TB in the other. Both studies showed increased sensitivity (from 60% to 88% and 53% to 80%) and decreased specificity (from 55% to 26% and 55% to 37%) with the addition of abnormal chest radiography. The pooled sensitivity in the studies of the combination of abnormal chest radiography plus the four-symptom screening rule (84.6%, 95% CI 69.7;92.9) was higher than that with the symptom screening rule alone (52.2%, 95% CI 38.0;66.0); however, specificity decreased (29.8%, 95% CI 26.3;33.6 vs 55.5%, 95% CI 51.8;59.2). The differences in sensitivity and specificity by screening type were both statistically significant.

Across studies, the median prevalence of TB among HIV-positive people on and not on ART was 1.5% (IQR: 0.6–3.5%) and 11.3% (IQR: 6.7–16.1%), respectively. When the prevalence of TB is 1.0%, the negative predictive value of the symptom screening rule is 99.3%, and addition of abnormal chest radiography increases it by 0.2%.

Subgroup Type of screening No. of studies

Pooled sensitivity (%) (95% CI)

Pooled specificity (%) (95% CI)

Negative predictive value for TB prevalence (%)

1 5 10 20

On ART

Symptom screening alone 7 51.0 (28.4;73.2) 70.7 (47.8;86.4) 99.3 96.5 92.8 85.2Symptom screening plus abnormal chest radiography

2 84.6 (69.7;92.9) 29.8 (26.3;33.6) 99.5 97.4 94.6 88.6

Not on ART

Symptom screening alone 15 89.3 (82.6;93.6) 27.2 (17.3;40.0) 99.6 98.0 95.8 91.1Symptom screening plus abnormal chest radiography

5 94.3 (76.2;98.8) 20.1 (7.6;43.8) 99.7 98.5 97.0 93.4

Pregnant women Symptom screening alone 4 27.1 (16.3;41.7) 82.4 (79.1;85.2) 99.1 95.6 91.1 81.9

Children Symptom screening alone 1 100 (76.8;100) 4.3 (1.8;8.7) 100 100 100 100


Bala

nce

of b

enefi

t vs

harm

Do the benefits outweigh the harms? Yes No Equal Uncertain

The anticipated desirable effect of screening is correct identification of PLHIV who do not have active TB and are thus eligible for TB preventive treatment (true negatives). The other desirable effect is correct identification of those with TB who would be confirmed by subsequent investigations (true positives). The anticipated undesirable effect is incorrect classification of an individual with TB as not having TB (false negatives), as this would lead to inappropriate treatment of active TB by a preventive treatment regimen. In addition, individuals who screen positive would have to undergo further investigations for TB when they are actually TB negative (false positives).

In the studies included in the review, the median prevalence of TB was 1.5% among PLHIV on ART. Accordingly, in a hypothetical population of 1000 PLHIV and at a TB prevalence of 1%, symptom screening alone would wrongly classify five TB patients as not having TB and being put on TB preventive treatment, while symptom screening plus abnormal chest radiography would wrongly put only two TB patients on preventive treatment.

At a TB prevalence of 1%, symptom screening alone would require TB investigations for 58 extra non-TB patients for every TB case identified. Similarly, when symptom screening plus abnormal chest radiography were used, the number of HIV-positive people requiring TB investigations would increase (87 extra non-TB patients for every TB case identified).

By adding abnormal chest radiography, more patients would have to undergo investigations when they don’t have TB. They might be lost to follow-up during investigations and miss an opportunity to be started on preventive treatment. Use of chest radiography could reduce concern of health workers about development of drug resistance.

Adults and adolescents on ART

Screening type Test accuracy Test resultsEffect per 1000 individuals screened Quality of

evidencePrevalence 1% Prevalence 5% Prevalence 10%

Symptom screening alone

Sensitivity (%): 51.0 (28.4;73.2)

Specificity (%): 70.7 (47.8;86.4)

True positive 5 (3–7) 26 (14–37) 51 (28–73) ㊉㊉◯◯ LowFalse negative 5 (3–7) 24 (13–36) 49 (27–72)

True negative 700 (473–855)

672 (454–821)

636 (430–778) ㊉㊉◯◯

LowFalse positive 290 (135–517) 278 (129–496) 264 (122–470)

Symptom screening plus abnormal chest radiography

Sensitivity (%): 84.6 (69.7;92.9)

Specificity (%): 29.8 (26.3;33.6)

True positive 8 (7–9) 42 (35–46) 85 (70–93) ㊉㊉㊉◯ ModerateFalse negative 2 (1–3) 8 (4–15) 15 (7–30)

True negative 295 (260–327)

283 (250–314)

268 (237–297) ㊉㊉㊉㊉

HighFalse positive 695

(663–30)667

(636–700)632

(603–663)


Evid

ence

of a

ccur

acy What is the overall

certainty of the evidence of test accuracy? Very low Low Moderate High No included studies

A systematic review was conducted, which identified two cross-sectional studies of the WHO-recommended four-symptom screening rule plus abnormal chest radiography. The studies involved 646 participants, of whom 39 (6.0%) had active TB. The quality of the evidence for true positive–false negatives was considered moderate because of serious imprecision, while that for true negative–false negative was high. In view of the moderate quality of the evidence of true positive–false negatives and taking into account the small number of studies, the overall quality of the evidence was considered low.

M

anag

emen

t effe

cts What is the overall

certainty of the evidence of effects of the management that is guided by the test results? Major uncertainty Minor uncertainty

The studies included in the review were not designed to assess the effects of management with different screening strategies on patient outcomes (e.g. active TB incidence, mortality, drug resistance).

The efficacy of preventive treatment might depend on confirmation of TB infection in an LTBI test.

Val

ues

Is there important uncertainty about or variation in how many people value the main outcomes? Important uncertainty

or variation No important

uncertainty or variation

Addition of abnormal chest radiography increases burden on patients. Patients may value greater certainty in excluding active TB.

Reso

urce

s re

quire

d

How large are the resource requirements (costs)? Greater resource

requirements Less resource

requirements Neither greater nor

less Varies Don’t know

More resources required, particularly if chest radiography is not available.Chest radiography would increase the number of HIV-positive people who undergo further investigations for TB.


Cos

t effe

ctiv

enes

sDoes the cost–effectiveness of the test favour the intervention or the comparison? Favours the





Cost–effectiveness could vary by region and health system infrastructure.

Equi

ty


Impact on health equity depends on the setting (e.g. availability of chest radiography: could increase or decrease equity).

Acc

epta

bilit

y Is the test acceptable to key stakeholders? No Yes Varies Don’t know

Depends on availability of resources and infrastructure (e.g. electricity, radiologists).

Feas

ibili

ty

Is the test feasible to implement? No Yes Varies Don’t know

Varies significantly, mainly by setting, health system infrastructure and workload of HIV clinics.


Summary of judgements Judgement Implications

Problem No Yes Varies Don’t know

Test accuracy Very inaccurate Inaccurate Accurate Very accurate Varies Don’t know


Certainty of the evidence of test accuracy

Very low Low Moderate High No incuded studies

Certainty of the evidence of effects of management

Major uncertainty Minor uncertainty

ValuesImportant



variability

Resources required Greater Neither greater nor less Less Varies Don’t know



nor the comparison

Favours the intervention Varies No incuded

studies

Equity Reduced Increased Varies Don’t know

Acceptability No Yes Varies Don’t know

Feasibility No Yes Varies Don’t know


ConclusionsWhat is the accuracy of WHO symptomatic screening plus abnormal chest radiography to exclude active TB in individuals with HIV on antiretroviral treatment (ART)?

Type of recommendation

Symptom screening alone ☐

Symptom screening plus chest radiography☒



Strong☐

Conditional☒

Recommendation Chest radiography may be offered to people living with HIV and on ART and preventive treatment be given to those with no abnormal radiographic findings. (Conditional recommendation, low-quality evidence)Remark: Chest radiography should not be a requirement for initiating preventive treatment.

Justification Overall, the GDG agreed that the screening rule based on four symptoms is very useful for ruling out active TB before providing preventive treatment to people living with HIV, regardless of whether they receive ART. It also noted the marginal potential benefits of adding abnormal chest radiography findings to the four-symptom screening rule. Moreover, increased use of chest radiography would pick up false-positives to the screening rule, so that more clients would be subjected to investigations for TB and other illnesses. Therefore, the GDG reiterated that chest radiography adds value only if it does not present a barrier for the provision of preventive treatment for people living with HIV.The GDG also noted that symptom screening with or without abnormal chest radiography findings would be acceptable to individuals and programme managers. Furthermore, the use of chest radiography could enhance the confidence of health care providers that active TB has been ruled out and reduce their concern for the development of drug resistance. The addition of chest radiography may incur costs to clients as well as inconvenience, as more clients will have to be investigated for TB and other diseases.


Although no study was found of the additive role of chest radiography in testing pregnant women, the GDG noted that pregnant women living with HIV could also benefit, as long as good clinical practices are observed to prevent any significant risk to the fetus. The GDG noted the paucity of data on the usefulness of the screening rule for children living with HIV. The single study showed that the symptom screening rule currently recommended for children with HIV performs well, but no study has been reported on the harm or challenges of the rule, such as resource requirements for implementation. Symptom-based screening is generally accepted by clients and is feasible in resource-constraint settings. Therefore, the GDG decided to make the same strong recommendation.


Addition of abnormal chest radiographic findings to the symptom screening rule would complicate logistics, increasing the cost, workload, infrastructure and availability of qualified staff. The GDG noted that chest radiography should not be a requirement or a barrier for initiating TB preventive treatment in people living with HIV because of the need for additional resources, in view of the marginal gain in negative predictive value. People living with HIV who have any of the four symptoms or abnormal chest radiographic findings may have active TB and should be investigated for TB and other diseases. Xpert MTB/RIF should be used as the initial diagnostic test. Other diseases that cause any of the four symptoms should be investigated in accordance with national guidelines and sound clinical practice. People living with HIV who present any of the four symptoms but in whom active TB is excluded by investigations may be considered for preventive treatment. The four-symptom screening method is recommended for all people living with HIV at every visit to a health facility or contact with a health worker. As combining chest radiography with symptom screening at every visit could represent a significant burden on the health system as well as on clients, it should be used only to exclude active TB before giving preventive treatment, with due respect for good clinical practice. The role of chest radiography in regular TB screening and its optimal frequency is uncertain. Local authorities should define its application and frequency on the basis of their local epidemiology, health infrastructure and resource availability. It is essential to ensure the availability of chest radiography and trained health care workers (e.g. radiologists) to implement the screening rule.


Research priorities • Performance and feasibility of the algorithms proposed in the present guidelines.• In particular, data on the screening rule for children and pregnant women.


GRADE tables Question: What is the performance of WHO-recommended four-symptom screening to exclude active TB in individuals with HIV?Population: Adults and adolescents with HIV on ART

Sensitivity 0.51 (95% CI: 0.28;0.73)

Specificity 0.71 (95% CI: 0.48;0.86) Prevalence 1% 5% 10%

Outcome

No. of studies;

no. of patients

Study design

Factors that may decrease the quality of evidence Effect per 1000 patients tested

Effect per 1000 patients tested

Effect per 1000 patients tested Test accuracy

quality of evidenceRisk of bias Indirectness Inconsistency Imprecision Publication

bias

Pre-test probability of

1%


5%

Pre-test probability of 10%

True positives (patients with active TB) 7

studies; 4640

patients

Cross-sectional (cohort type)

Not serious Not serious Serious1 Serious2 None3

5 (3–7) 26 (14–37) 51 (28–73)

㊉㊉◯◯ Low False negatives

(patients incorrectly classified as not having active TB)

5 (3–7) 24 (13–36) 49 (27–72)

True negatives (patients without active TB) 7

studies; 4640

patients

Cross-sectional (cohort type)

Not serious Not serious Serious1 Serious2 None3

700 (473–855) 672 (454–821) 636 (430–778)

㊉㊉◯◯ Low False positives

(patients incorrectly classified as having active TB)

290 (135–517) 278 (129–496) 264 (122–470)

From references 30–36 1 Significant heterogeneity for sensitivity and specificity. Downgraded by 1. 2 Wide confidence intervals. Downgraded by 1. 3 Possibility of publication bias not excluded, but not considered of sufficient concern to downgrade.


Question: What is the performance of combination of chest radiography and WHO-recommended four-symptom screening to exclude active TB in individuals with HIV?Population: Adults and adolescents with HIV on ART

Sensitivity 0.85 (95% CI: 0.70;0.93)

Specificity 0.30 (95% CI: 0.26;0.33) Prevalence 1% 5% 10%

Outcome No. of studies; no. of patients Study design

Factors that may decrease quality of evidence Effect per 1000 patients testedTest accuracy

Quality of evidenceRisk of bias Indirectness Inconsistency Imprecision Publication

bias

Pre-test probability

of 1%


5%


10%

True positives (patients with active TB)

2 studies; 646 patients

Cross-sectional

(cohort type) Not serious Not serious Not serious Serious1 None2

8 (7–9) 42 (35–46) 85 (70–93)

㊉㊉㊉◯ Moderate False negatives

(patients incorrectly classified as not having active TB)

2 (1–3) 8 (4–15) 15 (7–30)

True negatives (patients without active TB)

2 studies; 646 patients

Cross-sectional

(cohort type) Not serious Not serious Not serious Not serious None2

295 (260–327)

283 (250–314)

268 (237–297)

㊉㊉㊉㊉ High False positives

(patients incorrectly classified as having active TB)

695 (663–730)

667 (636–700)

632 (603–663)

From references 30 and 351 Imprecise estimate for sensitivity; downgraded by 1.2 Possibility of publication bias not excluded but not considered of sufficient concern to downgrade.


PICO3: What is the accuracy of symptomatic screening and/or chest radiography to exclude active TB in contacts of pulmonary TB cases without HIV in high TB incidence countries?

Population: Contacts of pulmonary TB cases who are HIV-negative. Background Active TB must be excluded before TB preventive treatment is provided. WHO recommends use of the symptom screening rule alone for excluding active TB in children aged < 5 years who are contacts of TB cases. For contacts in other age groups, however, there is no clear guidance on methods for excluding active TB, as these groups were not targets for LTBI treatment in high-TB incidence countries. In low-TB incidence countries, WHO currently recommends the combination of any TB symptoms and any chest radiography abnormality for excluding active TB before preventive treatment.

Intervention: Symptom screening and/or chest radiography.

Role of the test: Rule out active TB before providing preventive treatment.

Linked treatments: Screening negative ➞TB preventive treatment.

Anticipated outcomes:

True positive: Correct identification of an individual with active TB who should undergo further investigations.False negative: Incorrect identification of an individual with active TB as not having TB.True negative: Correct identification of an individual as not having active TB.False positive: Incorrect identification of an individual who should undergo further investigations who is actually TB negative.



Subgroups: Children.



Prob

lem


Globally in 2015, there were an estimated 10.4 million incident TB cases and 1.8 million TB deaths. In order to end the global TB epidemic, management of LTBI is critical, as stated in the WHO End TB Strategy. Active TB must be excluded before providing TB preventive treatment. A simple algorithm for excluding active TB is considered an essential component of programmatic LTBI management and could facilitate scaling-up of TB preventive treatment.

Test

acc

urac

y

How accurate is the test? Very inaccurate Inaccurate Accurate Very accurate Varies Don’t know

We updated a systematic review conducted in 2012 to determine the sensitivity and specificity of symptoms and chest radiography screening for active pulmonary TB in HIV-negative people and those of unknown HIV status. To illustrate how different screening and diagnostic algorithms are expected to perform in ruling out active TB, a simple model was constructed to compare six screening methods. The main findings are summarized in the tables below:

* No data could be obtained directly from the studies included in the systematic review; thus, the estimates were inferred from five studies of both chest radiography and symptom screening.

Performance of screening tools in a hypothetical population of 10 000 HIV-negative individuals at 2% TB prevalence

Algorithm No. of studies Sensitivity Specificity False negative at

screening

Negative predictive value after negative

screening

False positive at screening

Chest radiography: any abnormality 7 0.941 0.868 12 0.999 1294

Chest radiography: abnormality suggestive of TB

6 0.893 0.922 21 0.998 764

Any cough 10 0.627 0.775 75 0.990 2205

Cough ≥ 2–3 weeks 6 0.382 0.943 124 0.987 559

Any TB symptom 11 0.730 0.766 54 0.993 2303

Any TB symptom plus any chest radiography abnormality

* 1.00 0.701 0 1 2930


* No data could be obtained from the studies included in the systematic review; thus, the estimates were inferred from five studies of both chest radiography and symptom screening.

The sensitivity and negative predictive value of chest radiography screening are high, especially if any chest radiography abnormality is used. Symptom screening is less sensitive, resulting in a lower negative predictive value.In several studies, it was assumed that people without chest radiography abnormalities and without a minimum set of symptoms did not have active TB and that a positive culture may be only transient or due to laboratory cross-contamination or subclinical TB. This is a standard design in TB prevalence surveys.We identified only one study conducted among children < 5 years old (mean age, 19.2 months; standard deviation, 7.4). The sensitivity and specificity of abnormal chest radiography for TB (sensitivity, 55%, 95% CI 40;70; specificity, 89%, 95% CI 87;91) were higher than those of “persistent cough” (sensitivity, 45%, 95% CI 30;60; specificity, 84%, 95% CI 82;84). However, there was a high risk of selection bias, as the study included only children suspected of having TB from symptoms, contact history or known conversion to positive TST or IGRA.

Performance of the screening tools in a hypothetical population of 10 000 HIV-negative individuals at 5% TB prevalence

Algorithm No. of studies Sensitivity Specificity False negative at

screening

Negative predictive value after negative

screening

False positive at screening

Chest radiography: any abnormality 7 0.941 0.868 30 0.996 1254

Chest radiography: abnormality suggestive of TB

6 0.893 0.922 54 0.994 741

Any cough 10 0.627 0.775 187 0.975 2136

Cough ≥ 2-3 weeks 6 0.382 0.943 309 0.967 542

Any TB symptom 11 0.730 0.766 135 0.982 2233

Any TB symptom plus any chest radiography abnormality

* 1.00 0.701 0 1 2841


Bala

nce

of b

enefi

ts v

s ha

rmDo the benefits outweigh the harms? Yes No Equal Uncertain

One anticipated desirable effect of screening is correct identification of individuals who do not have active TB and are thus eligible for TB preventive treatment (true negatives). The other desirable effect is correct identification of those with TB that would be confirmed in subsequent investigations (true positives).The anticipated undesirable effect is incorrect classification of an individual with TB as not having TB (false negative), which would lead to inappropriate treatment of active TB by a preventive treatment regimen. In addition, individuals who screen positive have to undergo further investigations for TB when they are actually TB negative (false positive) and cannot be started on TB preventive treatment immediately. In a hypothetical population of 10 000 individuals and at a TB prevalence of 2%, use of any TB symptoms alone would wrongly classify 54 TB patients as not having active TB and they would be given TB preventive treatment. In contrast, use of any abnormal chest radiography finding would result wrongly in 12 TB patients being given preventive treatment. Use of the combination of any TB symptoms plus any chest radiography abnormal findings would result in no TB patients being given preventive treatment. At a TB prevalence of 2%, use of any TB symptoms alone would require TB investigations of 16 extra non-TB patients for every TB case identified, whereas use of any abnormal chest radiography finding would require TB investigations of 7 extra non-TB patients for every TB case identified. Use of the combination of any TB symptoms plus any chest radiography abnormal finding would increase the number of individuals requiring TB investigations to 15 extra non-TB patients for every TB case identified.

Cer

tain

ty o

f evi

denc

e of

test

acc

urac

y What is the overall certainty of the evidence of test accuracy? Very low Low Moderate High No included studies

The quality of the evidence for any chest radiography abnormality was judged as low–moderate, while that for any TB symptoms was very low. Furthermore, there was no direct evidence on the combination of any chest radiography abnormality plus any TB symptoms. Therefore, the overall certainty of the evidence is considered very low.


Cer

tain

ty o

f the

evi

denc

e of

man

agem

ent’s

effe

cts What is the overall

certainty of the evidence of effects of management guided by test results? Major uncertainty Minor uncertainty

The studies included were not designed to assess the effects of management with different screening strategies on patient outcomes (e.g. active TB incidence, mortality, drug resistance).

V

alue

s




Depends on health infrastructure and settings. Addition of abnormal chest radiography would increase burden on patients, although they might value an accurate test.

Reso

urce

s re

quire

d


requirements Less resource

requirements Neither greater nor

less Varies Don’t know

A systematic literature review was conducted for the previous LTBI guidelines, of studies published between 1981 and 2013 on the cost–benefit and cost–effectiveness of LTBI screening and treatment. In the 13 studies in which costs were expressed in US$, the cost of ruling out active TB in persons eligible for LTBI preventive treatment (including in most cases chest radiography, clinical evaluation and liver function tests) was US$ 28–188. Apart from a study conducted in India, the others were carried out in high-income and upper middle-income countries.Six studies on contacts of patients with active TB suggested that screening for and treatment of LTBI among contacts in general may save costs for the health care system and/or have a favourable incremental cost–effectiveness ratio. All the studies were conducted in low-TB incidence countries. Cost–effective data for various screening methods or algorithms were not available.


Cos

t effe

ctiv

enes

sDoes the cost–effectiveness of the test favour the intervention or the comparison? Favours the





Depends on the setting. It may be cost–effective in the long term by preventing development of drug-resistant TB.

Equi

ty


Acc

epta

bilit

y Is the test acceptable to key stakeholders? No Yes Varies Don’t know

Depends on setting and availability of chest radiography.

Feas

ibili

ty

Is the test feasible to implement? No Yes Varies Don’t know

Depends on setting and availability of chest radiography and human resources.


Summary of judgements Judgement Implications


Test accuracy Very inaccurate Inaccurate Accurate Very accurate Varies Don’t know


Certainty of the evidence of test accuracy

Very low Low Moderate High No included studies

Certainty of the evidence of effects on management

Major uncertainty Minor uncertainty

Values Important uncertainty or

variability


variability


Cost–effectivenessFavours the comparison


nor the comparison


studies





Conclusions What is the accuracy of symptomatic screening and/or chest radiography to exclude active TB in contacts of pulmonary TB cases without HIV in high TB incidence countries?

Type of recommendation

Any chest radiography abnormality ☐

Chest radiography abnormality suggestive of TB☐

Any cough☐

Cough ≥ 2–3 week☐

Any TBsymptom☐

Any TB symptom plus any chest radiography abnormality☒



Strong ☐

Conditional☒

Recommendation The absence of any symptoms and the absence of TB and of abnormal chest radiographic findings may be used to rule out active TB disease among HIV-negative household contacts aged ≥ 5 years and other at-risk groups before preventive treatment. (Conditional recommendation, very low-quality evidence)

Justification Overall, the GDG agreed that the potential benefits of screening for active TB with the combination of any chest radiography abnormality plus any TB symptoms outweighs the harm because of the reliability of this screening rule for excluding active TB before providing preventive treatment.The GDG also noted that symptom screening with or without the addition of abnormal chest radiography would be acceptable for individuals and programme managers. Furthermore, the use of chest radiography could enhance the confidence of health care providers that active TB has been ruled out and reduce their concern about development of drug resistance. However, the addition of chest radiography may incur costs to clients as well as inconvenience, as more clients will be investigated for TB and other diseases.



Contacts with abnormal chest radiography findings or TB symptoms must be followed up properly and investigated for TB and other diseases. Investigations should be performed in accordance with national guidelines and sound clinical practice. Contacts in whom active TB is excluded after investigations can be considered for preventive treatment. Chest radiography and trained health care workers (e.g. radiologists) must be available to implement the screening rule. Where chest radiography is not available, contacts should be screened for any TB symptoms. This would offer the highest sensitivity among the symptom screening rules, and its negative predictive value would remain high in most settings.


Research priorities Evidence for the accuracy and feasibility of the recommended screening algorithm under programme conditions. Household models to improve the effectiveness and efficiency of intervention delivery.Studies of cost–effectiveness of screening rules.Strategies to save costs and improve feasibility (e.g. use of mobile chest radiography).


GRADE tables Question: What is the accuracy of symptomatic screening and/or chest x-ray to exclude active TB in contacts of pulmonary TB cases without HIV in high TB incidence countries?Index test: any abnormality in chest radiography| Reference test: Sputum culture and/or smear Place of testing: Triage Test–treatment pathway: chest radiography positive ➞ confirmatory test (mycobacterial culture or GeneXpert) ➞ anti-TB chemotherapy (6–9 months of antibiotics)


Factors that may decrease quality of evidence Effect per 100 000 Sensitivity: 0.94 (95% CI: 0.86;0.98)Specificity: 0.87 (95% CI: 0.80;0.92)

Quality of evidenceRisk of bias Indirectness Inconsistency Imprecision Publication bias

True positives (patients with active TB) 7 studies;

251 410 patients

Cross-sectional

(cohort type) Serious1 Not serious2 Not serious3 Not serious4 None5

Prevalence (2%): 1 882 (1 716;1 954)Prevalence (5%): 4 705 (4 290;4 885 ㊉㊉㊉◯

ModerateFalse negatives (patients incorrectly classified as not having active TB)

Prevalence (2%) : 118 (46;284)Prevalence (5%): 295 (115;710)

True negatives (patients without active TB) 7 studies;

251 410 patients

Cross-sectional

(cohort type) Serious1 Not serious2 Not serious3 Not serious4 None5

Prevalence (2%) : 85 064 (78 106;89 866)Prevalence (5%): 82 460 (75 715;87 115) ㊉㊉㊉◯

Moderate False positives (patients incorrectly classified as having active TB)

Prevalence (2%) : 12 936 (8 134;19 894)Prevalence (5%): 12 540 (7 885;19 285)

Studies included: references 37,41,44,46-491 Limitations in study design (see QUADAS-2): High risk of selection bias in one study (37). In all studies, less than half the participants received the reference standard; accuracy was calculated under the assumption that those

who did not receive the reference standard were culture- and/or smear-negative (no active TB). 2 Indirectness (see QUADAS-2): Some concern about applicability of reference standard in two studies. No downgrading. 3 Inconsistency: Little heterogeneity in sensitivity or specificity (from visual inspection of 95% CIs). 4 Imprecision: Precise estimates for sensitivity and specificity. 5 Publication bias: Not applicable (the evidence for publication bias in studies of diagnostic test accuracy is very limited).


Question: What is the accuracy of symptomatic screening and/or chest x-ray to exclude active TB in contacts of pulmonary TB cases without HIV in high TB incidence countries?Index test: Any symptom| Reference test: Sputum culture and/or smear Place of testing: Triage Test–treatment pathway: Symptom positive ➞ confirmatory test (mycobacterial culture or GeneXpert) ➞ anti-TB chemotherapy (6–9 months’ antibiotics)


Factors that may decrease quality of evidence Effect per 100 000 Sensitivity: 0.73 (95% CI: 0.64;0.80) Specificity: 0.77 (95% CI: 0.61;0.87)

Quality of evidenceRisk of bias Indirectness Inconsistency Imprecision Publication

bias

True positives (patients with active TB) 11 studies;

357 609 patients

Cross-sectional (cohort type) Very serious1 Not serious2 Not serious3 Not serious4 None5

Prevalence (2%): 1 460 (1 282;1 608) Prevalence (5%): 3 650 (3 205;4 020) ㊉㊉◯◯

Low False negatives (patients incorrectly classified as not having active TB)

Prevalence (2%): 540 (392;718) Prevalence (5%): 1 350 (980;1 795)

True negatives (patients without active TB) 11 studies;

357 609 patients

Cross-sectional (cohort type) Very serious1 Not serious2 Serious3 Serious4 None5

Prevalence (2%): 74 970 (60 074;85 260)Prevalence (5%): 72 675 (58 235;82 650) ㊉◯◯◯

Very low False positives (patients incorrectly classified as having active TB)

Prevalence (2%): 23 030 (12 740;37 926) Prevalence (5%): 22 325 (12 350;36 765)

From references 37–471 Limitations in study design (see QUADAS-2): High risk of selection bias in one study (37) and unclear risk of bias for the reference standard in two studies. In 9 of the 11 studies, less than half the participants received the reference

standard; accuracy was calculated under the assumption that those who did not receive the reference standard were culture- and/or smear-negative (no active TB). 2 Indirectness (see QUADAS-2): no major concern for applicability. 3 Inconsistency: moderate heterogeneity for sensitivity and significant heterogeneity for specificity (based on visual inspection of 95% CIs); downgrading on specificity. 4 Imprecision: precise estimates for sensitivity and imprecise estimate for specificity. 5 Publication bias: not applicable (the evidence for assessing publication bias in studies of diagnostic test accuracy is very limited).


PICO 4: Could interferon-gamma release assays be used as an alternative to tuberculin skin tests to identify individuals most at risk of progression from LTBI to active TB in high TB incidence settings?

Problem Assess use of IGRA as an alternative to TST for identifying individuals at greatest risk of progression from LTBI to active TB in high-TB incidence settings.

Background There is no gold standard for the diagnosis of LTBI. TST and IGRA indirectly identify TB infection by detecting memory T-cell response signifying the presence of host sensitization to Mycobacterium tuberculosis antigens. They are generally deemed to be acceptable but imperfect tests.WHO currently recommends that IGRA should not replace TST in high-TB incidence countries on the basis of a systematic review that showed similar performance in predicting development of active TB and its high cost and technical complexity. Either IGRA or TST can be used to test for LTBI in high-income and upper–middle-income countries with an estimated TB incidence < 100 per 100 000. Because of the global shortage of RT23 purified protein derivative, however, many countries are having difficulty in accessing it. The availability of an alternative test, IGRA, may facilitate scaling-up of programmatic LTBI management. Although sensitivity and specificity are usually used to evaluate the diagnostic accuracy of a test, there is no gold standard test for LTBI, and preventive treatment is meant to prevent the development of active TB. Therefore, the performance of tests for LTBI is better assessed from their predictive utility for development of active TB. The primary effect measure of interest is the relative risk ratio for TB among test-positives and test-negatives, which will be compared for TST and IGRA.

Option: IGRA

Comparison: TST

Main outcomes: Incidence of active TB.

Setting: High-TB incidence countries (estimated TB incident rate ≥ 100 per 100 000 population).



Prob

lem

Is the problem a priority? No Yes Varies Don’t Know

Currently, LTBI testing is not required before provision of preventive treatment in high-TB incidence countries. It can identify individuals who would benefit most from LTBI treatment and is used in some high-incidence countries. Lack of availability of TST because of the global shortage of purified protein derivative has been cited as a barrier to scaling-up of programmatic management of LTBI. The availability of an alternative test, IGRA, may facilitate scaling-up.


Bal

ance

of e

ffect

sDo the benefits outweigh the harm? Yes No Equal Uncertain

Five relevant studies of IGRA and TST in high-TB incidence countries were identified (N = 7769). All were prospective cohort studies of participants who received both TST and IGRA. Two were conducted in India and three in South Africa. The populations studied were people living with HIV, pregnant women, adolescents, health care workers and household contacts. The RRs for test positives and test negatives were estimated for each test and pooled across studies. The pooled RR estimate was 1.49 for TST (95% CI 0.79;2.80, 5 studies, I2 = 64.4%) and 2.03 (95% CI 1.18;3.50, 5 studies, I2 = 49.6%) for IGRA. Although the pooled effect estimate for IGRA was slightly higher and the heterogeneity lower than for TST, the 95% CIs around the effect estimates overlapped and were imprecise.

There was little evidence for specific at-risk populations. Two studies were conducted in people living with HIV, and the pooled estimates were imprecise.

Cer

tain

ty o

f evi

denc


Val

ues




No evidence retrieved.

TST IGRA

Population Pooled RR I2 (p value) Pooled RR I2 (p value)

All populations (5 studies)

1.49 (0.79;2.80)

64.4% (0.024)

2.03 (1.18;3.50)

49.6%(0.094)

People living with HIV (2 studies)

1.64 (0.24;11.18)

77.4% (0.035)

4.07 (0.18;92.72)

78.7% (0.030)


Reso

urce

s re

quire





Varies Don’t Know

A systematic review of studies of cost–effectiveness was conducted for the previous LTBI guidelines, which covered 39 studies published up to 2013. Cost inputs adjusted for currency and inflation varied widely among studies. The cost of a TST for detecting LTBI varied from US$ 1.3 in a study in Uganda to an average of US$ 31.5 in studies in the United Kingdom. Detection of LTBI with a IGRA test cost from US$ 22.5 in a study in Mexico to an average of US$ 97.1 in studies in the United Kingdom.

C

ost e

ffect

iven

ess





Uncertain Varies No included studies

A systematic review (50) of 10 studies with a decision-analytical model for comparing the cost–effectiveness of IGRAs with that of TST in high-risk groups: child contacts, immunocompromised people and recent arrivals from high-TB incidence countries. One study of child contacts was conducted in South Africa and the others in low-TB incidence countries. The study in South Africa showed that providing preventive treatment without testing is most cost–effective among children aged 0–2 years. In children aged 3–5 years, an IGRA after a negative TST saved slightly more life-years, but saving one additional life year costed at least US$ 233 000. Six cost evaluations were conducted among immunocompromised people (including people living with HIV) in Japan and the USA. Five studies showed that IGRA is more cost–effective than TST. In one study of patients taking immunosuppressive medicine, neither TST nor IGRA screening was more cost–effective than treatment without testing. These results depend on the performance of TST and IGRA assumed in the models, and the studies generally assumed higher sensitivity and/or specificity of IGRA for diagnosing LTBI. A systematic review conducted for the previous guidelines, which was updated in June 2017, covered five studies of TST and IGRA screening in adult contacts. None was conducted in high-TB incidence countries. Two indicated that the TST alone was more cost–effective than IGRA alone; two found that IGRA was more cost–effective than TST alone but less cost–effective than sequential TST-IGRA. One study indicated that both strategies were better than no LTBI screening or treatment.

Very limited data from high-TB incidence countries.Results of cost–effectiveness studies in low-incidence countries may not be generalizable to high-incidence countries.

Equi

ty

What would be the impact on health equity? Reduced Increased Varies Don’t Know

No evidence retrieved. The provision of more options generally increases equity; however, if the cost of the test is borne by patients, use of IGRA might be a greater barrier and might decrease equity.


Acc

epta

bilit

yIs the intervention acceptable to key stakeholders? No Yes Varies Don’t Know

No evidence retrieved. Acceptability varies, particularly by resource availability. Although IGRA is likely to be largely acceptable to clinicians, its higher cost and requirement for sophisticated laboratory infrastructure may limit its acceptability to programmes. Both IGRA and TST have been used widely in many countries and are accepted.

Feas

ibili

ty

Is the intervention feasible to implement? No Yes Varies Don’t Know

Depends on the availability of resources and tests.IGRA: Phlebotomy is required, particularly for very young children, and sophisticated laboratory infrastructure, technical expertise and expensive equipment are required. TST: Can be performed in the field; training for intradermal injection, reading and interpretation are required, and there are frequent stock-outs due to global shortage.Both tests have been available for many years and are used widely in many countries.






studies


variability


variability




nor the comparison

Favours the intervention Uncertain No included

studies





ConclusionsCould interferon-gamma release assays be used as an alternative to tuberculin skin tests to identify individuals most at risk of progression from LTBI to active TB in high TB incidence settings?


Against☐



Strong☒

Conditional☐

Recommendation Either a TST or IGRA can be used to test for LTBI. (Strong recommendation, very low-quality evidence)Remark: The availability and affordability of the tests will determine which will be chosen by clinicians and programme managers. Neither TST nor IGRA can be used to diagnose active TB disease nor for diagnostic workup of adults suspected of having active TB.

Justification The GDG concluded that the comparison of TST and IGRA in the same population does not provide strong evidence that one test should be preferred over the other for predicting progression to active TB disease. The GDG noted that TST may require significantly fewer resources than IGRA and may be more familiar to practitioners in resource-constrained settings; however, recurrent global shortages and stock-outs of TST reduce its use in scaling up programmatic management of LTBI. The GDG also noted that equity and access could affect the choice and type of test used. The preferences of clients and programmes are, however, affected by several factors, such as the requirement for sophisticated laboratory infrastructure (e.g. for IGRA) and possible additional costs for clients (e.g. for travel) and programmes (e.g. for building and testing). The GDG strongly recommended the two tests as equivalent options, with relatively similar advantages and disadvantages.The GDG stressed that the global shortage of TST should be addressed urgently and called for more investment into research on novel tests for LTBI with better predictive value. The GDG cautioned that imperfect performance of these tests can lead to false-negative results, particularly for young children and immunocompromised individuals such as people living with HIV. The GDG noted the importance of the tests for identifying recent conversion from a negative to a positive result, particularly among contacts of people with pulmonary TB, which is good practice for initiating TB preventive treatment. Nevertheless, recent studies among health care workers tested serially for LTBI in the USA showed that conversions from negative to positive and reversions from positive to negative are more commonly identified with IGRA than with TST. Thus, sound clinical judgement must be used in interpreting the results of these tests when used serially. The GDG recommended that LTBI testing should not be a requirement for initiating TB preventive treatment for people living with HIV and child household contacts aged < 5 years, particularly in countries with a high TB incidence, given that clear benefits outweigh the risks. HIV-negative infant and child household contacts aged < 5 years and people living with HIV who have a negative LTBI test should be assessed case by case for their individual risk of exposure to TB and the added advantage of receiving preventive treatment.




The GDG noted that the availability and affordability of the tests could determine which LTBI test is used. Other considerations include the structure of the health system, feasibility of implementation and infrastructure requirements. The incremental cost-effectiveness of IGRAs and TSTs appears to be influenced mainly by their accuracy. Bacille Calmette-Guérin (BCG) vaccination plays a decisive role in reducing the specificity of TST, leading the choice towards IGRA-only strategies. The GDG noted, however, that the impact of BCG vaccination on the specificity of TST depends on the strain of vaccine used, the age at which the vaccine is given and the number of doses administered. When BCG is given at birth, as is the case in most parts of the world, it has a variable, limited impact on TST specificity. Therefore, the GDG agreed that a history of BCG vaccination has a limited effect on interpretation of TST results later in life; hence, BCG vaccination should not be a determining factor in selecting a test. IGRAs are more costly and more technically complex to perform than the TST. Operational difficulties should be considered in deciding which test to use. For example, IGRA requires a phlebotomy, which can be difficult, particularly in very young children, laboratory infrastructure, technical expertise and expensive equipment; however, only a single visit is required to obtain a result (although patients may have to make a second visit to learn the result). TST is less costly and can be performed in the field, but it requires a cold chain, two health care visits and training in intradermal injection, reading and interpretation.


Research priorities New tests with better predictivity for progression from LTBI to active TB disease than current tests.Predictive performance of both tests in various at-risk populations.Cost–effectiveness studies under different conditions of burden and subgroups (e.g. children, people living with HIV).


GRADE table: Studies that conducted head to head evaluations of the TST and IGRA (N=5) Review question: Among persons at high risk of LTBI who are not treated with tuberculosis preventive therapy, which test (e.g. TST or IGRA) when positive, can best identify individuals most at risk of progression? SR Outcome: The predictive utility of the tuberculin skin test vs. the commercial interferon-gamma release assays for progression to active tuberculosis Patients/population: Longitudinal studies of adults and children without active TB at baseline not given preventive therapySetting: Community cohorts, individuals attending outpatient clinics (e.g. HIV-positive people), individuals participating in RCTs, household contacts; all in high-incidence countries Index test: TSR (RT23 purified protein derivative or purified protein derivative-S) and/or commercial blood-based IGRAs (QFT-GIT or T.SPOT.-TB) Importance: Longitudinal studies on the predictive value of a positive IGRA in TB high-incidence countries (≥ 100/100 000) are still emerging. It is important to determine whether IGRA can be used as a replacement for the widely used TST.Reference standard: All diagnoses of incident active TB (microbiologically confirmed or not)Studies: Any longitudinal study design (e.g. prospective or retrospective cohort) in TB high-incidence countries, regardless of immunological status (e.g. HIV-infected or not) or BCG status. Average follow-up should be for at least 1 year but can be either active or passive.

No. of studies (no. of individuals) Design

Quality Effect Quality (GRADE) Importance

Risk of bias Inconsistency Indirectness Imprecision Relative (pooled) Absolute effect

A. SR OUTCOME: PROGRESSION TO ACTIVE TB IN UNTREATED INDIVIDUALS

5 (N = 7675 for TST, 7641 for IGRA) (51–55)

Prospective cohort

Serious risk of bias

(A1) (-1)

Serious inconsistency (TST)I2 = 64.4%,

Serious inconsistency (IGRA)I2 = 49.6%

(A2) (-1)

Not serious

(A3)

Serious imprecision(TST)

No serious imprecision (IGRA)

(A4) (-1)

TSTRR = 1.49 (CI: 0.79;2.80) I2 = 64.4%

IGRARR = 2.03(CI: 1.18;3.50)I2 = 49.6%

TST10 more per 1000 (4 fewer to 37 more)

IGRA15 more per 1000 (3 more to 36 more)

Very low㊉◯◯◯

Critical

B. SR OUTCOME (SUBGROUP ANALYSIS): PROGRESSION TO ACTIVE TB IN IMMUNOCOMPROMISED PEOPLE (INCLUDES HIV AND OTHER IMMUNOSUPPRESSIVE CONDITIONS)

2 (N = 725 for TST, 710 for IGRA) (52, 54)

Prospective cohort of HIV-infected women pre- and post-delivery on ARTProspective cohort of HIV-infected individuals


(B1) (-1)

Serious inconsistency (TST) I2 = 77.4%

Serious inconsistency (IGRA) I2 = 78.7%

(B2) (-1)

Serious indirectness

(B3) (-1)

Very seriousimprecision for both TST and IGRA

(B4) (-2)

TSTRR = 1.64(CI: 0.24;11.18)

IGRARR = 4.07 (CI: 0.18;92.72)


IGRA149 more per 1000 (40 fewer to 4438 more)


Critical


No. of studies (no. of individuals) Design

Quality Effect Quality (GRADE) Importance

Risk of bias Inconsistency Indirectness Imprecision Relative (pooled) Absolute effect

C. SR OUTCOME (SUBGROUP ANALYSIS) : PROGRESSION TO ACTIVE TB AMONG CONTACTS OF TB CASES

1 (N = 1511 for TST, 1498 for IGRA) (55)

Prospective cohort of household contacts


(C1) (-1)

Not assessed; single study

(C2)

Serious Indirectness

C3 (-1)

Serious imprecision

C4 (-1)

TSTRR, single study = 1.31 (CI: 0.85;2.04)

IGRA RR, single study = 1.87 (CI: 1.12;3.11)




Critical

D. SR OUTCOME (SUBGROUP ANALYSIS): PROGRESSION TO ACTIVE TB AMONG TB HEALTH CARE WORKERS

1 (N = 195 for TST, 189 for IGRA) (53)

Prospective cohort of health care workers


(D1) (-1)


(D2)


D3 (-1)

Very serious imprecision

D4 (-2)


IGRARR, single study = 3.10 (CI: 0.13;75.04)

TST6 fewer per 1000 (9 fewer to 82 more)

IGRA(difference cannot be computed)


Critical

E. SR OUTCOME (SUBGROUP ANALYSIS): PROGRESSION TO ACTIVE TB AMONG ADOLESCENTS IN A HIGH-INCIDENCE SETTING

1 (N = 5244 for both tests) (51)

Prospective cohort of adolescents


(E1) (-1)


(E2)


E3 (-1)

No serious imprecision

E4


IGRARR, single study = 2.89 (CI: 1.55;5.41)

TST9 more per 1000 (2 more to 21 more)



Critical

*Absolute risk: estimated by applying the RR estimate to the risk in the test negatives.


Notes to the GRADE summary tableOverall quality: One point was removed from all the studies because none were RCTs. The lowest quality score achievable is 1 out of 4; no minus scores are given.Quality assessment: Based on the relative effect measure (RR or IRR) for both TST and IGRA. Studies not marked down if estimates for both tests scored high on a specific GRADE quality item.Other study quality considerations: Newcastle–Ottawa scale quality items were considered when assessing the risk of bias. One point is removed if there is at least one concern.

A1: Risk of bias is possible, including selection bias, incorporation bias, ascertainment bias and publication bias. Methods for ascertaining TB included microbiological methods, but not all incident TB cases were confirmed definitively by culture. Publication bias not formally assessed but expected to be likely. Several large prospective studies are under way or unpublished, and their results were not included in this analysis; however, additional results are not expected to change the overall conclusions of this review.A2: Serious unexplained inconsistency of RR estimate for TST. Points removed for serious inconsistency in either estimate.A3: Although there were few studies included, they involved a range of populations, including adults and children, immunocompromised people and TB contacts, and provided direct evidence for these groups.A4: Serious imprecision of RR estimate for TST. Lower limit of 95% CI indicates lack of predictivity. Points removed if serious imprecision was identified in either estimate.

B1: Risk of bias is possible, including selection bias, incorporation bias, ascertainment bias and publication bias. Incorporation bias could not be ruled out for the cohort of antepartum and postpartum women, because relevant information was not available; moreover, there was concern about selection. The reference standards used in the ART cohort study did not include index tests, and the assessors were not blinded to baseline TST results in patient records. Methods for ascertaining TB included microbiological methods, but not all incident TB cases were definitively diagnosed. Publication bias was not formally assessed but is expected to be likely. Several large prospective studies are under way or are unpublished, and their results were not included in this analysis; however, additional results are not expected to change the overall conclusions of this review.B2: Serious unexplained inconsistency for RR estimates for both TST and IGRA. B3: This pooled estimate is based on only two studies: one on HIV-infected people on ART with a median CD4+ of approximately 250, and one on HIV-infected antepartum and postpartum women. No direct evidence for treatment of naive patients or HIV-infected patients with high CD4 counts or other sub-populations of HIV-infected individuals (e.g. children). B4: Very serious imprecision of RR estimates for both TST and IGRA. The 95% CIs are wide and indicate both significant predictive performance and lack of predictive utility. The studies had few events.

C1: Risk of bias is possible, including selection bias, incorporation bias (could not be assessed because of lack of information) and publication bias. Publication bias was not formally assessed but was expected to be likely. Several large prospective studies are under way or are unpublished, and their results were not included in this analysis; however, additional results are not expected to change the overall conclusions of this review.C2: Inconsistency not assessed.C3: This single study comprised household case contacts in a high-incidence country. No direct evidence for other subpopulations of case contacts. C4: TST effect estimates seriously imprecise. Lower limit of 95% CI indicates lack of predictive utility.

D1: Risk of bias is possible, including selection bias, ascertainment bias (microbiological tests not used to diagnose TB), incorporation bias and publication bias. Publication bias was not formally assessed but was expected to be likely. Several large prospective studies are under way or are unpublished, and their results were not included in this analysis; however, additional results are not expected to change the overall conclusions of this review.D2: Inconsistency not assessed.D3: This single study comprised health care workers at a primary health care clinic. No direct evidence for other subpopulations of health care workers or all health care settings. D4: IGRA and TST effect estimates very seriously imprecise; 95% CIs are wide and indicate both significant predictive performance and lack of predictive utility.

E1: Risk of bias is possible, including selection bias, ascertainment bias (inclusion of index tests in methods for ascertaining incident TB) and publication bias. Publication bias was not formally assessed but is expected to be likely. Several large prospective studies are under way or are unpublished, and their results were not included in this analysis; however, additional results are not expected to change the overall conclusions of this review. E2: Inconsistency not assessed. E3: This single study comprised adolescents in a high-incidence setting. No direct evidence for other subpopulations of children or adolescents. E4: No serious imprecision: few events with large sample size.


PICO 5: Should 3-month daily rifampicin plus isoniazid (3RH) be offered as a preventive treatment option for children and adolescents <15 years of age as an alternative to 6 or 9 months isoniazid (INH) monotherapy in high TB incidence countries?

Problem Children and adolescents < 15 years with LTBI and at high risk for active TB disease.

Background Treatment of LTBI can reduce the risk of reactivation by 60–90%. WHO currently recommends two approaches for the management of LTBI, based on TB incidence and income. For high-TB incidence countries, WHO recommends isoniazid preventive therapy for people living with HIV and children aged < 5 years who are household contacts of people with TB. The recent WHO guidelines provide several treatment options for use in high- or upper–middle-income countries with low TB incidence. A previous systematic review suggested that the efficacy of a 3-month regimen of daily rifampicin plus isoniazid is similar to that of daily isoniazid regimens.

Option: 3 months’ daily rifampicin + isoniazid (3RH).

Comparison: 6 or 9 months’ isoniazid monotherapy.

Main outcomes: Incidence of active TB, mortality, adverse events, treatment completion rate, drug-resistant TB.




Prob

lem


Uptake of LTBI treatment is still suboptimal: only 38% of people living with HIV were newly enrolled in care in 2015 and 7.1% of child household contacts < 5 years started on preventive treatment. A systematic review (56) showed that failure to complete treatment accounts for a large loss in the cascade of care for LTBI management. Shorter regimens may improve completion rate and facilitate scaling-up of LTBI treatment in high-TB incidence countries.


Bala

nce

of e

ffect

sDo the benefits outweigh the harms? Yes No Uncertain Equal

A systematic review covered one RCT and two observational studies. In the RCT, no cases of clinical TB disease were reported. Significantly fewer children given 4RH than those given 9H developed new radiography abnormalities suggestive of TB. In the same study, higher treatment adherence rate and fewer adverse events were observed in children given 3–4RH than in those given 9H.

Cer

tain

ty o

f evi

denc


Although the quality of the evidence was low, data on adult populations support the benefits of 3RH.

Val

ues




We conducted an online survey to solicit the values and preferences of individuals affected by the recommendations (Annex 3). Data were available from 142 respondents, of whom 59 had at least one child. The respondents were asked to rate the importance of each attribute of the LTBI treatment regimen on a five-point scale on which 5 is “very important” and 1 is “not important”. 90–100% of the respondents with children rated the following attributes as “very important” or “important” for their children: shorter duration, fewer side-effects, fewer visits to the clinic, easy to swallow and less frequent intake. Fewer respondents (78.0%) rated “no need for direct observed therapy (DOT)” as “very important” or “important”.

Outcome 3–4RH 6H/9H Relative effect (RR) (95% CI) Difference (95% CI)

Incidence of active TB (1 RCT)

26/220 (11.8%) 48/200 (24.0%) RR 0.492 (0.318–0.762)

122 fewer per 1000 (from 57 fewer to 164 fewer)

Adverse events (1 RCT) 27/650 (4.2%) 25/200 (12.5%) RR 0.332

(0.197–0.559)

83 fewer per 1000 (from 55 fewer to 100 fewer)

Adverse events (1 observational study)

1/220 (0.5%) 5/264 (1.9%) RR 0.24 (0.03–2.04)

14 fewer per 1000 (from 18 fewer to 20 more)

Completion rate (1 RCT) 220/238 (92.4%) 200/232 (86.2%) RR 1.07

(1.01–1.14)

60 more per 1000 (from 9 more to 121 more)

Completion rate(1 observational study)

48/72 (66.7%) 29/105 (27.6%) RR 2.41 (1.70–3.43)

389 more per 1000 (from 193 more to 671 more)


Reso

urce

s re

quire





Varies Don’t Know

No evidence retrieved. Treatment is shorter with 3RH than 6H/9H. Use of 3RH would require fewer resources, particularly because the drug combination is already being used for treatment of active TB.

Cos

t effe

ctiv

enes

s






No evidence retrieved. Fewer resources required with 3RH, while its effectiveness is greater because of higher completion rate and safer profile. Cost–effectiveness favours 3RH in studies in adult populations.

Equi

ty

What would be the impact on health equity? Reduced Increased Varies Don’t Know

No evidence retrieved. The availability of more options would increase equity in accessing health services.

Acc

epta

bilit

y

Is the intervention acceptable to key stakeholders? No Yes Varies Don’t Know

No evidence retrieved.


Feas

ibili

tyIs the intervention feasible to implement? No Yes Varies Don’t Know

Co-administration of rifampicin with protease inhibitors is not recommended. Rifampicin is known to significantly lower plasma concentrations of dolutegravir, and the dosing schedule might have to be increased to to twice daily, but there are very few studies and limited clinical experience with this combination (57).

Drug interactions preclude its co-administration with protease inhibitors or nevirapine (e.g. infants born to HIV-positive mothers receiving nevirapine). Little concern about drug interactions in HIV-negative child contacts.

Summary of judgements

Judgement Implications

Problem No Yes Varies Don’t Know



studies

ValuesImportant



variability

Resources required Greater Neither greater nor less Less Varies Don’t Know


Favours neither the intervention or

the comparison


studies

Equity Reduced Increased Varies Don’t Know

Acceptability No Yes Varies Don’t Know

Feasibility No Yes Varies Don’t Know


ConclusionsShould 3-month daily rifampicin/isoniazid (3RH) be offered as preventive treatment option for children and adolescents < 15 years of age as an alternative to 6 or 9 months of isoniazid monotherapy in high-TB incidence countries?


Against☐



Strong☒

Conditional☐

Recommendation Rifampicin plus isoniazid daily for 3 months should be offered as an alternative to 6 months of isoniazid monotherapy as preventive treatment for children and adolescents aged < 15 years in countries with a high TB incidence. (Strong recommendation, low-quality evidence)

Justification The GDG unanimously agreed that the benefits of 3RH outweigh the harm, given its safer profile, higher completion rate than with isoniazid monotherapy and the availability of child-friendly fixed-dose combinations of rifampicin and isoniazid.The GDG noted that, although the quality of the evidence was low, data on adult populations also support the benefits of 3RH. A systematic review of RCTs on preventive treatment options conducted in 2014 showed that the efficacy and the risk for hepatotoxicity are similar for 3RH and isoniazid monotherapy. The GDG noted that use of 3RH would require fewer resources, given the shorter duration of treatment, which would reduce the number of clinic visits required. It also suggested that the initial cost of use of 3RH would be low, as it is already being used for treatment of active TB. The GDG agreed that cost–effectiveness favours 3RH because of the higher completion rate, safer profile and fewer resources required. The GDG also noted that, although direct evidence for the cost–effectiveness of 3RH in children is limited, the cost–effectiveness of shorter preventive treatment including 3RH is supported by a body of evidence in adult populations. The GDG agreed that there is no important uncertainty or variability in clients’ values and preferences. It also agreed that the acceptability of 3RH is high, given its shorter duration and long use by health care workers for treatment of active TB disease.



The GDG strongly encouraged use of paediatric fixed-dose combinations of rifampicin and isoniazid for children, as they will increase acceptability and feasibility. It also noted that 3RH should be prescribed with caution to people living with HIV who are on ART because of potential drug–drug interactions; the regimen cannot be co-administered with protease inhibitors or nevirapine. The GDG further emphasized the importance of surveillance systems for rifampicin-resistance TB.


Research priorities Further research on reliable methods for excluding active TB among children.


GRADE table Question: Should 3-month daily rifampicin/isoniazid (3RH) be offered as preventive treatment option for children and adolescents < 15 years of age as an alternative to 6 or 9 months’ isoniazid monotherapy in high-TB incidence countries?

Overall quality: lowQuality assessment No. of patients Effect

Quality ImportanceNo. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other

considerations

3–4-month daily rifampicin

+ isoniazid

6–9-month isoniazid

monotherapy

Relative (95% CI)

Absolute (95% CI)

“RADIOLOGICAL” TB DISEASE: (58) (FOLLOW UP: 3–7 YEARS TO 7–11 YEARS; ASSESSED WITH: CHEST RADIOGRAPHY)

1 RCT Serious1 Not serious Serious2 Not serious None 26/220 (11.8%)

48/200 (24.0%)

RR 0.492 (0.318–0.762)

122 fewer per 1000

(from 57 fewer to 164 fewer)

㊉㊉◯◯ Low Critical

MORTALITY

0 Cannot be estimated – Important

ADVERSE EVENTS: (58) (FOLLOW UP: 3–7 YEARS TO 7–11 YEARS; ASSESSED BY RECOGNITION OF SYMPTOMS AND ELEVATED LIVER ENZYMES)

1 RCT Very serious1,3 Not serious Serious4 Not serious None 27/650

(4.2%) 25/200 (12.5%)

RR 0.332 (0.197–0.559)

83 fewer per 1000


㊉◯◯◯ Very low Critical

ADVERSE EVENTS: (59) (FOLLOW UP: MEDIAN 97–197 DAYS; ASSESSED WITH: LIVER TOXICITY TEST AND CLINICAL)

1 Observational Serious5 Not serious Serious4 Serious6 None 1/220 (0.5%) 5/264 (1.9%)

RR 0.24 (0.03–2.04)

14 fewer per 1000

(from 18 fewer to 20 more)

㊉◯◯◯Very low Critical

COMPLETION RATE: (58) (FOLLOW UP: 3–7 YEARS TO 7–11 YEARS)9


200/232 (86.2%)

RR 1.07 (1.01–1.14)

60 more per 1000

(from 9 more to 121 more)



Quality assessment No. of patients Effect


considerations

3–4-month daily rifampicin

+ isoniazid

6–9-month isoniazid

monotherapy

Relative (95% CI)

Absolute (95% CI)

COMPLETION RATE: (60) (ASSESSED FROM: COMPLETING > 80% OF TREATMENT WITHOUT INTERRUPTION OF > 2 MONTHS)

1 Observational studies Serious5 Not serious Not serious Serious8 None 48/72

(66.7%) 29/105 (27.6%)

RR 2.41 (1.70–3.43)

389 more per 1000



DRUG-RESISTANT TB


From references 58–601 Although there was a risk of selection bias, the characteristics of the two groups were similar. Patients with poor compliance were not included in the analysis of treatment outcomes. Downgraded by one level. 2 There was no clinical disease. The outcome reported was new radiography findings suggestive of possible active disease. No comparison with 6H. Downgraded by one level.3 High risk of detection bias because of lack of blinding. The RH group included participants enrolled during the second period, whose characteristics were different; they were not randomized between the RH group and the 9H

group. Downgraded by two levels.4 No comparison with 6H. Downgraded by one level.5 Risk of bias because of non-comparability of the two groups. Downgraded by one level.6 Low event rate and wide 95% CI. Downgraded by one level.7 Lack of blinding. Medication adherence test performed at home by parents. Although there was a risk of selection bias, the characteristics of the two groups were similar. Downgraded by one level.8 Wide 95% CI. Downgraded by one level.9 Adherence rates reported; compliance considered poor if no medication was detected in urine strips, if patients did not return for follow-up visits or if they were lost to follow-up. Poor compliance was considered non-completion

in the analysis.


PICO 6: Should 3-month weekly rifapentine and isoniazid be offered as an alternative regimen to isoniazid monotherapy for treatment of LTBI in high TB incidence countries?

Problem Individuals with LTBI who are at high risk for active TB disease. Background Treatment of LTBI can reduce the risk for reactivation by 60–90%. WHO currently recommends two approaches for the management of LTBI, based on TB incidence and income. For high-TB incidence countries, WHO recommends isoniazid preventive therapy for people living with HIV and children aged < 5 years who are household contacts of people with TB. The recent WHO guidelines provide several treatment options for high- or upper–middle-income countries with low TB incidence. A previous systematic review suggested that the efficacy of the weekly regimen was similar to daily isoniazid regimens, with higher treatment completion rates and a safer profile.

Option: 3-month weekly rifapentine and isoniazid (3HP).

Comparison: Isoniazid monotherapy.

Main outcomes: Incidence of active TB, mortality, adverse events, treatment completion, drug resistance.




Prob

lem


Uptake of LTBI treatment is still suboptimal, with only 38% of people living with HIV newly enrolled in care and 7.1% of child household contacts < 5 years started on preventive treatment in 2015. A systematic review (56) showed that failure to complete treatment accounts for a large loss in the cascade of care for LTBI management. A previous review of LTBI treatment options (61) suggested that the efficacy of the weekly regimen was similar to that of daily isoniazid, with higher treatment completion rates and a safer profile. Therefore, 3HP could significantly facilitate scaling-up of LTBI treatment in high-TB incidence countries.

Bal

ance

of e

ffect

s

Do the benefits outweigh the harm? Yes No Equal Uncertain

We conducted a systematic review with the following subgroup analyses: adults with HIV, adults without HIV, and children and adolescents. Regardless of subgroup, there was no significant difference in the incidence of active TB in participants given 3HP and 6-months’ isoniazid (6H) or 9-months’ isoniazid (9H). 3HP was associated with higher completion rates (RR, 1.09–1.25) and fewer adverse events (RR, 0.63–0.88) than 6 or 9 months’ isoniazid monotherapy in all subgroups. In a comparison of 3HP and continuous isoniazid, the trial showed no significant difference in TB incidence in the intention-to-treat analysis; however, a per-protocol analysis showed a lower rate of TB or deaths among participants given continuous isoniazid rather than 3HP. 3HP was associated with significantly fewer adverse events than continuous isoniazid (RR 0.20, 95% CI 0.12;0.32).

Cer

tain

ty o

f evi

denc


The overall quality of the evidence was considered high for the comparison between 3HP and 6/9H in adults with HIV, moderate in adults without HIV and in children and adolescents. It was considered moderate for the comparison of 3HP with continuous isoniazid in adults with HIV.


Val

ues




We conducted an online survey to solicit the values and preferences of individuals affected by the recommendations (Annex 3). Data were available from 142 respondents, including 10 reported as HIV-positive. The respondents were asked to rate the importance of each attribute of the LTBI treatment regimen on a five-point scale on which 5 is “very important” and 1 is “not important”. More than 90% of the respondents considered the following attributes of preventive treatment to be very important or important: shorter duration, fewer side-effects, fewer visits to the clinic and fewer pills. Fewer respondents rated “less frequent intake” and “no need for DOT” as very important or important (77.3% and 74.4%, respectively). Similarly, while less than 80% of the participants rated “no need for DOT” as very important or important for their children, all the other attributes were rated as very important or important by 90–100%.

Re

sour

ces

requ

ired





Varies Don’t Know

No evidence retrieved. Implementation of 3HP would require more resources, particularly if it is to be given under DOT.

Cos

t effe

ctiv

enes

s






In a cost–effective analysis of 3HP in the USA (62), the cost was assumed to be US$6.00 per 900-mg dose of rifapentine and US$ 0.05 per dose of isoniazid. Over 20 years, 3HP given by DOT would cost the health system US$ 8861 more per TB case prevented and US$ 1879 more per quality-adjusted life year gained than 9H. From the social perspective, 3HP given by DOT was considered cost-saving.The study also found that, if adherence to self-administered 3HP is maintained at levels achieved by DOT, 3HP given by self-administration would cost less than 9H from both a health system and a social perspective.

Varies in different settings depending on cost of the drug and mode of administration (DOT or self-administration).


Equi

tyWhat would be the impact on health equity? Reduced Increased Varies Don’t Know

No evidence retrieved. The availability of more options is generally considered to increase equity.

Acc

epta

bilit

y

Is the intervention acceptable to key stakeholders? No Yes Varies Don’t Know

No evidence retrieved. Acceptability varies by risk group and setting, including mode of administration (self-administration or DOT).

Feas

ibili

ty

Is the intervention feasible to implement? No Yes Varies Don’t Know

In all the RCTs in the review, 3HP was administered under DOT. Non-inferiority of self-administered 3HP with or without text reminders for DOT was not established in the overall study population. Non-inferiority was achieved in a subgroup analysis among participants in the USA.Studies of pharmacokinetics suggest that rifapentine can be co-administered with efavirenz or raltegravir without dose adjustment. A study of the pharmacokinetics of co-administration of dolutegravir and 3HP was terminated prematurely because of the development of an influenza-like syndrome and elevated liver transaminases in two of four participants. Data on co-administration of rifapentine with other antiretroviral drugs are limited; however, as rifapentine is a potent inducer of P450 enzymes and the P-glycoprotein transport system, interactions with some antiretroviral drugs are expected. No significant interaction is expected when co-administered with abacavir, emtricitabine, tenofovir-DF, lamivudine or zidovudine. Potential interactions are expected with nevirapine and protease inhibitors. In addition, although co-administration has not been studied, rifapentine is expected to significantly reduce plasma concentrations of tenofovir alafenamide, etravirine and rilpivirine.

Feasibility depends on settings and risk groups and is mainly affected by the mode of delivery and drug interactions. The GDG noted unpublished data that suggested the effectiveness and acceptability of self-administration.






studies

ValuesImportant



variability




nor the comparison


studies





ConclusionsShould 3-month weekly rifapentine and isoniazid be offered as an alternative regimen to isoniazid monotherapy for treatment of LTBI in high-TB incidence countries?


Against☐



Strong☐

Conditional☒

Recommendation Rifapentine and isoniazid weekly for 3 months may be offered as an alternative to 6 months of isoniazid monotherapy as preventive treatment for both adults and children in countries with a high TB incidence. (Conditional recommendation, moderate-quality evidence)

Justification The GDG agreed unanimously that the benefits of 3HP outweigh the harm, given the similar preventive efficacy, safer profile and higher completion rate of 3HP than isoniazid monotherapy. The GDG noted that use of 3HP would require more resources, particularly if 3HP is administered by DOT. One cost–effectiveness study conducted in the USA suggested that 3HP may be more cost-saving than 9-months isoniazid. There was consensus in the GDG that the cost–effectiveness of 3HP depends mainly on the cost of the drug and mode of administration, which would affect the costs to patients and health systems.There was consensus in the GDG that the acceptability of 3HP varies by risk group and setting, due mainly to the mode of administration (self–administration or DOT). The GDG considered that adding 3HP as an alternative to isoniazid would provide more options and hence increase equity.


The GDG recognized the lack of data on use of 3HP in pregnant women and children < 2 years and stressed the need for data on these populations.


The GDG noted that 3HP can be self-administered. Evidence from an RCT suggests that adherence to self-administered treatment of 3HP is not inferior to DOT. There is little further evidence on use of the 3-month regimen of weekly rifapentine plus isoniazid. The GDG noted that a requirement for DOT could be a significant barrier to the implementation.3HP should be prescribed with caution to people living with HIV who are on ART because of potential drug–drug interactions. The GDG noted that the 3HP can be administered to patients receiving efavirenz-based antiretroviral regimens without dose adjustment, according to a study of pharmacokinetics. Administration of rifapentine with raltegravir was found to be safe and well tolerated. Rifapentine-containing regimens should not be administered with dolutegravir until more information becomes available. The GDG urged further studies on the pharmacokinetics of 3HP with a variety of drugs, particularly ART.


The GDG stressed the importance of recording and reporting on the provision and completion of TB preventive treatment according to standardized indicators, in order to monitor progress in implementation.

Research priorities • Value of self-administration of 3HP.• Studies of pharmacokinetics with a variety of drugs, particularly ART.• Use of 3HP in pregnant women and children < 2 years old.


GRADE tablesQuestion: Should a 3-month regimen of weekly rifapentine plus isoniazid be offered as an alternative regimen to daily isoniazid monotherapy for treatment of LTBI in high-TB incidence countries? Population: Adults with HIVComparison: 6 or 9 months of isoniazid monotherapy

Overall quality: highQuality assessment No. of patients Effect

Quality ImportanceNo. of studies

Study design Risk of bias Inconsistency Indirectness Imprecision Other

considerations3-month weekly RPT+isoniazid

6 or 9 months isoniazid

Relative (95% CI)

Absolute (95% CI)

ACTIVE TB

2 RCTs Not serious Not serious Not serious1 Serious2 None 26/534 (4.9%) 28/520 (5.4%)

RR 0.733 (0.234–2.295)

14 fewer per 1000


㊉㊉㊉◯ Moderate Critical

ALL-CAUSE MORTALITY

2 RCTs Not serious Not serious Not serious1 Serious2 None 23/535 (4.3%) 30/513 (5.8%)

RR 0.746 (0.438–1.270)

15 fewer per 1000

(from 16 more to 33 fewer)

㊉㊉㊉◯ Moderate Important

ANY ADVERSE EVENT (GRADE III OR IV)

2 RCTs Serious3 Not serious Not serious1 Not serious None 39/535 (7.3%) 59/513 (11.5%)

RR 0.627 (0.426–0.921)

43 fewer per 1000



HEPATOTOXICITY

2 RCTs Not serious4 Not serious Not serious1 Not serious None 8/535 (1.5%) 30/513 (5.8%)

RR 0.256 (0.118–0.553)

44 fewer per 1000

(from 26 fewer to 52

fewer)

㊉㊉㊉㊉ High Critical

DRUG-RESISTANT TB

2 RCTs Not serious Not serious Not serious1 Very serious5 None 3/534 (0.6%) 1/520

(0.2%)

RR 2.001 (0.259–15.436)

2 more per 1000


㊉㊉◯◯ Low Important


COMPLETION RATE

2 RCTs Not serious Not serious Not serious1 Not serious None 497/534 (93.1%)

397/520 (76.3%)

RR 1.255 (1.014–1.553)

195 more per 1000



From references 63 and 641 Although one of the trials was conducted in low-TB incidence countries, this is unlikely to affect the relative effect of RPT/isoniazid compared with isoniazid monotherapy. Not downgraded. 2 95% CIs of both relative and absolute effect indicate appreciable benefit and harm with 3HP. 3 Both trials were open-label, which may have introduced bias in ascertainment of adverse events. 4 Although the trials were open-label, this is unlikely to affect detection of hepatotoxicity, which is usually done by objective measurement (i.e. blood tests). Not downgraded. 5 Very low event rates. Upper limit of 95% CIs of both relative and absolute effect include appreciable harm with 3HP. Downgraded by two levels.

Question: Should a 3-month regimen of weekly rifapentine plus isoniazid be offered as an alternative regimen to daily isoniazid monotherapy for treatment of LTBI in high-TB incidence countries? Population: Adults with HIVComparison: Continuous isoniazid monotherapy

Overall quality: moderateQuality assessment No. of patients Effect


Study design

Risk of bias Inconsistency Indirectness Imprecision Other

considerations3-month weekly RPT+isoniazid

Continuous isoniazid

Relative (95% CI)

Absolute (95% CI)

ACTIVE TB

1 RCT Not serious Not serious Not serious Serious1 None 24/328 (7.3%) 8/164 (4.9%) RR 1.500

(0.689–3.265)

24 more per 1000



ALL-CAUSE MORTALITY

1 RCT Not serious Not serious Not serious Serious1 None 17/328 (5.2%) 8/164 (4.9%) RR 1.063

(0.468–2.410)

3 more per 1000



ANY ADVERSE EVENTS (GRADE III OR IV)

1 RCT Serious2 Not serious Not serious Not serious None 21/328 (6.4%) 53/164 (32.3%)

RR 0.198 (0.124–0.317)

259 fewer per 1000


fewer)



HEPATOTOXICITY

1 RCT Not serious3 Not serious Not serious Not serious None 5/328 (1.5%) 46/164

(28.0%) RR 0.054

(0.022–0.134)

265 fewer per 1000


fewer)


DRUG-RESISTANT TB

1 RCT Not serious Not serious Not serious Very

serious4 None 2/328 (0.6%) 1/164 (0.6%) RR 1.000 (0.091–10.948)

0 fewer per 1000


㊉㊉◯◯ Low Important

COMPLETION RATE

1 RCT Not serious Not serious Not serious Not serious None 314/328

(95.7%) 99/164 (60.4%)

RR 1.586 (1.398–1.799)

354 more per 1000

(from 240 more to 482

more)


From reference 631 95% CIs of both relative and absolute effect indicate appreciable benefit and harm with 3HP.2 The trial was open-label, which may have introduced bias in ascertainment of adverse events. 3 Although the trial was open-label, this is unlikely to affect detection of hepatotoxicity, which is usually done by objective measurement (i.e. blood tests). Not downgraded. 4 Very low event rates. The upper limits of 95% CIs of both relative and absolute effect indicate appreciable harm with 3-month weekly RPT and isoniazid. Downgraded by two levels.


Question: Should a 3-month regimen of weekly rifapentine plus isoniazid be offered as an alternative regimen to daily isoniazid monotherapy for treatment of LTBI in high-TB incidence countries? Population: Adults without HIVComparison: 6 or 9 months of isoniazid monotherapy

Overall quality: moderate Quality assessment No. of patients Effect


Study design Risk of bias Inconsistency Indirectness Imprecision Other

considerations3-month

RPT+isoniazid6 or 9 months’

isoniazidRelative

(95% CI)Absolute (95% CI)

ACTIVE TB

1 RCT Not serious Not serious Serious1 Not serious2 None 7/3986 (0.2%) 15/3745 (0.4%)

RR 0.438 (0.179–1.074)

2 fewer per 1000



ALL-CAUSE MORTALITY

1 RCT Not serious Not serious Serious1 Not serious3 None 31/3986 (0.8%)

39/3759 (1.0%)

RR 0.740 (0.462–1.183)

3 fewer per 1000



ANY ADVERSE EVENTS (GRADE III OR IV)


244/3759 (6.5%)

RR 0.873 (0.733–1.040)

8 fewer per 1000



HEPATOTOXICITY

1 RCT Not serious5 Not serious Serious1 Not serious None 18/4040

(0.4%) 103/3759

(2.7%) RR 0.163

(0.099–0.268)

23 fewer per 1000



DRUG-RESISTANT TB

1 RCT Not serious Not serious Serious1 Not serious3 None 1/3986 (0.0%) 2/3745 (0.1%) RR 0.470 (0.043–5.179)

0 fewer per 1000




COMPLETION RATE

1 RCT Not serious Not serious Serious1 Not serious None 3273/3985 (82.1%)

2585/3745 (69.0%)

RR 1.190 (1.159–1.221)

131 more per 1000



From reference 651 No study provided a comparison with 6 months of isoniazid. The study included 2.7% HIV-positive participants. Although the trial was conducted in low-TB incidence countries, this is unlikely to affect the effect of RPT/isoniazid

as compared with isoniazid monotherapy. Downgraded by one level. 2 Although the 95% CI of the RR is wide, there were few events, and the CI of the absolute effect is narrow. The result also met pre-stated non-inferiority margin. Not downgraded. 3 Although the 95% CI of the RR is wide, there were few events, and the CI of the absolute effect is narrow. Not downgraded. 4 The open-label design of the trial may have introduced ascertainment bias. Downgraded by one level.5 Although the trial was open-label, this is unlikely to affect detection of hepatotoxicity, which is usually done by objective measurement (i.e. blood tests). Not downgraded.

Question: Should a 3-month regimen of weekly rifapentine plus isoniazid be offered as an alternative regimen to daily isoniazid monotherapy for treatment of LTBI in high-TB incidence countries? Population: Children and adolescentsComparison: 6 or 9 months’ isoniazid

Overall quality: moderateQuality assessment No. of patients Effect


considerations3-month

RPT+isoniazid6 or 9 months

isoniazidRelative


ACTIVE TB


3/434 (0.7%)

RR 0.132 (0.007–2.542)

6 fewer per 1000



ALL-CAUSE MORTALITY


2/493 (0.4%)

RR 0.183 (0.009–3.802)

3 fewer per 1000



ANY ADVERSE EVENT (GRADE III OR IV)

1 RCT Serious4 Not serious Serious1 Not serious3 None 7/539 (1.3%)

8/493 (1.6%)

RR 0.875 (0.320–2.396)

2 fewer per 1000




HEPATOTOXICITY

1 RCT Not serious5 Not serious Serious1 Not serious None 0/539 (0.0%)

0/493 (0.0%)

Cannot be estimated

0 fewer per 1000

(from 4 fewer–4 more)


DRUG-RESISTANT TB


COMPLETION RATE

1 RCT Not serious Not serious Serious1 Not serious None 415/471 (88.1%)

351/434 (80.9%)

RR 1.089 (1.030–1.153)

72 more per 1000


㊉㊉㊉◯Moderate Critical

From reference 661 No study provided a comparison with 6 months of isoniazid. Although the trial was conducted in low-TB incidence countries, this is unlikely to affect the relative effect of RPT/isoniazid as compared with isoniazid monotherapy.

Downgraded by one level. 2 Although the 95% CI of the RR is wide, there were few events, and the CI of the absolute effect is narrow. The result also met pre-stated non-inferiority margin. Not downgraded. 3 Although the 95% CI of the RR is wide, there were few events, and the CI of the absolute effect is narrow. Not downgraded. 4 The open-label design of the trial may have introduced ascertainment bias. 5 Although the trial was open-label, this is unlikely to affect detection of hepatotoxicity, which is usually done by objective measurement (i.e. blood tests). Not downgraded.


PICO 7: Should preventive treatment be recommended for contacts of patients with multidrug-resistant or rifampicin-resistant TB?Problem Contacts of people with MDR or rifampicin-resistant TB. Background

People who have been in close contact with a TB case and who have become infected with M. tuberculosis are at high risk of progression to active disease, especially in the first 2 years after infection. Although TB preventive treatment is part of many TB control programmes, isoniazid monotherapy is unlikely to be effective in contacts of MDR-TB cases. In 2014, a guideline development group convened by WHO reviewed the evidence on use of preventive treatment of contacts of people with MDR-TB but could not make a recommendation because of the limited quality of the evidence. Rifampicin-resistant TB is considered a proxy for MDR-TB.

Option: Tailored preventive treatment.

Comparison: No treatment (only follow-up observation).

Main outcomes: Incidence of active TB disease, incidence of MDR-TB. mortality, adverse events.

Setting: High- and low-TB incidence countries.




Prob

lem


Drug-resistant TB continues to threaten global TB control, remains a major public health concern and poses a global health security risk. An estimated 580 000 people developed MDR or rifampicin-resistant TB in 2015, and 250 000 people died as a result (67). Prevention of MDR-TB would reduce the global burden and also address demands from individuals to be protected against development of MDR-TB.

Bal

ance

of e

ffect

s

Do the benefits outweigh the harm? Yes No Equal Uncertain

We conducted a systematic review of the effectiveness of preventive treatment for contacts of patients with MDR or rifampicin-resistant TB. The review covered 10 studies with control groups, of which five found no TB case in either group. The table below summarizes the results after exclusion of studies with < 20 participants who completed preventive TB treatment and those on isoniazid monotherapy.

Common adverse events included gastrointestinal symptoms, muscle or joint pain, headache, dizziness and hepatitis. In four studies, ≥ 50% of participants experienced at least one adverse event. Bamrah et al. (69) reported no serious adverse events, defined as hospitalization or irreversible morbidity, attributable to fluoroquinolone-based preventive treatment. The median proportion of participants who discontinued treatment because of adverse events in all studies was 5.1% (IQR 1.9–30.2%). No study reported preventive treatment for contacts of rifampicin-resistant TB.

Outcomes Reference Intervention Control Relative effect (OR) (95% CI)

Difference (95% CI) Follow-up

Incidence of active TB 68 2/41 (4.9%) 13/64

(20.3%) 0.20 (0.04–0.94) 154 fewer per 1000 (273 fewer to 36 fewer) 30 months

69 0/93 (0%) 3/15 (20%) 0.02 (0.00–0.39) 200 fewer per 1000 (403 fewer to 3 more) 36 months

70 0/21 (0%) 0/10 (0%) – 0 more per 1000 (138 fewer to 138 more) 12 months

71 0/30 (0%) 0/166 (0%) – 0 more per 1000 (45 fewer to 45 more) ≤ 9 years

Incidence of MDR-TB 69 0/93 (0%) 3/15 (20%) 0.02(0.00–0.39) 200 fewer per 1000 (403

fewer to 3 more) 36 months

70 0/21 (0%) 0/10 (0%) – 0 more per 1000 (138 fewer to 138 more) 12 months

71 0/30 (0%) 0/166 (0%) – 0 more per 1000 (45 fewer to 45 more) ≤ 9 years


Cer

tain

ty o

f evi

denc


The overall quality of the evidence was very low because of very serious risks of bias and imprecision. In the study by Trieu et al. (71), active TB was ascertained during follow-up by checking cases identified in the TB registry. A meta-analysis was not conducted because of the heterogeneity of the drugs used.

V

alue

s




Minimal uncertainty

We conducted an online survey to solicit the values and preferences of individuals affected by the recommendations (Annex 3). Data were available from 142 respondents. More than 80% of the respondents reported that they would strongly or somewhat prefer to receive preventive treatment or give it to their children if they were exposed to someone with MDR-TB disease in the household. The reasons for not preferring preventive treatment included: limited evidence on preventive treatment for MDR-TB and concern about side-effects and development of drug resistance.

There is uncertainty about the characteristics of respondents.

Reso

urce

s re

quire

d





Varies Don’t know


Cos

t effe

ctiv

enes

sDoes the cost–effectiveness of the intervention favour the intervention or the comparison? Favours the





Providing preventive treatment could be cost–effective by preventing MDR-TB cases in settings with low transmission of MDR-TB. In settings with high risk of MDR-TB transmission, the potential benefit may wane and the cost–effectiveness becomes uncertain. The need for drug susceptibility testing, regimens used, risk of re-infection and adverse events could also affect cost–effectiveness.

Equi

ty


Acc

epta

bilit

y

Is the intervention acceptable to key stakeholders? No Yes Varies Don’t know

Some national or clinical guidelines already recommend preventive treatment for contacts of MDR-TB (72–74). Preventive treatment could be acceptable, particularly to patients and health care workers. The intervention may not be acceptable in some settings, particularly to programme managers for fear of development of XDR-TB and little experience in using TB preventive treatment for drug-susceptible TB.

Feas

ibili

ty

Is the intervention feasible to implement? No Yes Varies Don’t know






studies


variability

Minimal uncertainty


variability




nor the comparison


studies





ConclusionsShould preventive treatment be recommended for contacts of patients with MDR or rifampicin-resistant TB?


Against☐



Strong☐

Conditional☒

Recommendation In selected high-risk household contacts of patients with multidrug-resistant tuberculosis, preventive treatment may be considered based on individualised risk assessment and a sound clinical justification. (Conditional recommendation, very low-quality evidence)

Remarks The preventive treatment should be individualized after a careful assessment of the intensity of exposure, the certainty of the source case, reliable information on the drug resistance pattern of the source case and potential adverse events. The preventive treatment should be given only to household contacts at high risk (e.g. children, people receiving immunosuppressive therapy and people living with HIV). The drugs should be selected according to the drug susceptibility profile of the source case. Confirmation of infection with LTBI tests is required. This recommendation must not affect on-going placebo-controlled clinical trials of MDR-TB contacts on ethical grounds. The results of such clinical trials are crucial for updating this recommendation. Strict clinical observation and close monitoring for the development of active TB disease for at least 2 years are required, regardless of the provision of preventive treatment.

Justification Overall, the GDG judged that the potential benefits of targeted preventive treatment for MDR-TB contacts based on individual risk assessments outweigh the harm but acknowledged uncertainty about the efficacy of the intervention due to the lack of RCTs. It also noted that provision of preventive treatment for MDR-TB contacts would be acceptable, particularly to patients and health care workers. The GDG stressed that treatment should be given to selected individuals after a careful risk assessment, including intensity of exposure, certainty of the source case, reliable information on the drug resistance pattern of the index case and potential adverse events. It should be given only to household contacts at high risk (e.g. children, people on immunosuppressive therapy and people living with HIV). Confirmation of infection by LTBI testing is required before individualized treatment is initiated.



Close monitoring and treatment adherenceClose monitoring of adverse events and adherence to treatment is essential. The types of adverse events depend on the drugs used. Common adverse events associated with each drug are listed in the Companion handbook to the WHO guidelines for the programmatic management of drug-resistant tuberculosis (75). Adverse effects should be monitored according to the WHO framework for monitoring and managing the safety of drugs against active TB (76). The GDG reiterated that strict clinical observation and close monitoring for active TB disease based on sound clinical practice and national guidelines for at least 2 years is required, regardless of the provision of preventive treatment. Consideration should also be given to interactions with antiretroviral, immunosuppressant and other drugs when providing TB preventive treatment.

Informed consent As the recommendation is based on very low-quality evidence, clients must be given detailed information about the benefits and harms of the preventive treatment and asked for explicit informed consent. In view of the uncertainty about the balance of benefit to harm, informed consent, preferably in writing, is required, based on the local context and practice in similar situations.


Selection of drug regimenThe regimen of preventive treatment of MDR-TB contacts should be based on reliable information on the drug resistance profile of the source case. Later-generation fluoroquinolones (e.g. levofloxacin and moxifloxacin) are considered to be important components of a preventive treatment regimen unless the strain of the source case is resistant to them. Although there has been concern about the use of fluoroquinolones in children because retardation of cartilage development was shown in animals, similar effects have not been demonstrated in humans. There is limited evidence for the duration of treatment, and this should be based on clinical judgement. The regimens used in the studies conducted so far were given for 6, 9 and 12 months.

Resources and feasibilityFor a programmatic approach, all the necessary resources should be in place, including for quality-assured testing for drug susceptibility, the necessary medications and a system for close monitoring of harm and adverse events. The feasibility of providing preventive treatment should be carefully assessed according to the availability of resources and the history and status of preventive treatment for drug-susceptible TB.


Research priorities • Adequately powered RCTs to update the recommendation on preventive treatment for MDR-TB contacts. • Effectiveness and safety of preventive treatment for MDR contacts under operational conditions. • Further evidence on risk of progression to active TB among MDR contacts to understand the benefits of preventive treatment.


GRADE tableQuestion: Should preventive treatment be recommended for contacts of patients with MDR or rifampicin-resistant TB?

Overall quality: very lowQuality assessment No. of patients Effect

Quality ImportanceNo. of studies Study design Risk of

bias Inconsistency Indirectness Imprecision Other considerations

Preventive treatment No treatment Relative


INCIDENCE OF ACTIVE TB DISEASE (BOTH DRUG-SUSCEPTIBLE AND DRUG-RESISTANT TB)

4 Observational Very serious1 Not serious Not serious Very

serious2 None

2/41 (4.9%) 13/64 (20.3%) 0.20 (0.04–0.94)3

154 fewer per 1000 (273 fewer to 36

fewer)


0/93 (0%) 3/15 (20%) 0.02 (0.00–0.39)4


more)

0/21 (0%) 0/10 (0%) –5

0 more per 1000 (138

fewer to 138 more)

0/30 (0%) 0/166 (0%) –60 more per

1000 (45 fewer to 45 more)

INCIDENCE OF MDR-TB

37 Observational Very serious1 Not serious Not serious Very

serious2 None

0/93 (0%) 3/15 (20%) 0.02 (0.00–0.39)4


more)

㊉◯◯◯ Very low Critical0/21 (0%) 0/10 (0%) –5

0 more per 1000 (138

fewer to 138 more)

0/30 (0%) 0/166 (0%) –60 more per

1000 (45 fewer to 45 more)


MORTALITY

0 No evidence available

Cannot be estimated – Important

ADVERSE EVENTS

0 No evidence available

Cannot be estimated – Critical

DEVELOPMENT OF DRUG RESISTANCE

0 No evidence available Important

From references 68–71. Five studies in which fewer than 20 participants completed preventive TB treatment were excluded, as was a study by Kristi (77), in which only isoniazid monotherapy was used.1 Risk of bias in selection of the control group, and confounders were not adjusted for in any study. Downgraded by two levels.2 Small sample sizes and wide 95% CIs. Downgraded by two levels.3 Reference 684 Reference 695 Reference 706 Reference 717 The study by Shaaf et al. (68) was excluded as the incidence of MDR-TB was not reported.


References

1. Mandalakas AM, Hesseling AC, Gie RP, Schaaf HS, Marais BJ, Sinanovic E. Modelling the cost-effectiveness of strategies to prevent tuberculosis in child contacts in a high-burden setting. Thorax. 2013;68(3):247–55.

2. Kasambira TS, Shah M, Adrian PV, Holshouser M, Madhi SA, Chaisson RE, et al. QuantiFERON-TB Gold In-Tube for the detection of Mycobacterium tuberculosis infection in children with household tuberculosis contact. Int J Tuberc Lung Dis. 2011;15(5):628–34.

3. Kenyon TA, Creek T, Laserson K, Makhoa M, Chimidza N, Mwasekaga M, et al. Risk factors for transmission of Mycobacterium tuberculosis from HIV-infected tuberculosis patients, Botswana. Int J Tuberc Lung Dis. 2002;6(10):843–50.

4. Klausner JD, Ryder RW, Baende E, Lelo U, Williame JC, Ngamboli K, et al. Mycobacterium tuberculosis in household contacts of human immunodeficiency virus type 1-seropositive patients with active pulmonary tuberculosis in Kinshasa, Zaire. J Infect Dis. 1993;168(1):106–11.

5. Bokhari SY, Ahmad A, Shaikh MY, Ahmad I. A study of tuberculosis contacts. J Pak Med Assoc. 1987;37(2):48–52.

6. Biraro IA, Kimuda S, Egesa M, Cose S, Webb EL, Joloba M, et al. The Use of Interferon Gamma Inducible Protein 10 as a Potential Biomarker in the Diagnosis of Latent Tuberculosis Infection in Uganda. PLoS One. 2016;11(1):e0146098.

7. Rutherford ME, Nataprawira M, Yulita I, Apriani L, Maharani W, van Crevel R, et al. QuantiFERON(R)-TB Gold In-Tube assay vs. tuberculin skin test in Indonesian children living with a tuberculosis case. Int J Tuberc Lung Dis. 2012;16(4):496–502.

8. Tornee S, Kaewkungwal J, Fungladda W, Silachamroon U, Akarasewi P, Sunakorn P. Risk factors for tuberculosis infection among household contacts in Bangkok, Thailand. The Southeast Asian journal of tropical medicine and public health. 2004;35(2):375–83.

9. Zelner JL, Murray MB, Becerra MC, Galea J, Lecca L, Calderon R, et al. Bacillus Calmette-Guerin and isoniazid preventive therapy protect contacts of patients with tuberculosis. Am J Respir Crit Care Med. 2014;189(7):853–9.

10. Tuberculosis Research Centre, Indian Council of Medical Research. Risk of tuberculosis among contacts of isoniazid-resistant and isoniazid-susceptible cases. Int J Tuberc Lung Dis. 2011;15(6):782–8.

11. Radhakrishna S, Frieden TR, Subramani R, Santha T, Narayanan PR, Indian Council of Medical R. Additional risk of developing TB for household members with a TB case at home at intake: a 15-year study. Int J Tuberc Lung Dis. 2007;11(3):282–8.

12. Narain R, Nair SS, Rao GR, Chandrasekhar P. Distribution of tuberculous infection and disease among households in a rural community. Bull World Health Organ. 1966;34(4):639–54.

13. WHO Tuberculosis Chemotherapy Centre. An investigation of household contacts of open cases of pulmonary tuberculosis amongst the Kikuyu in Kiambu, Kenya. Bull World Health Organ. 1961;25(6):831–50.

14. Andrews RH, Devadatta S, Fox W, Radhakrishna S, Ramakrishnan CV, Velu S. Prevalence of tuberculosis among close family contacts of tuberculous patients in South India, and influence of segregation of the patient on early attack rate. Bull World Health Organ. 1960;23:463–510.


15. Loudon RG, Williamson J, Johnson JM. An analysis of 3,485 tuberculosis contacts in the city of Edinburgh during 1954–1955. Am Rev Tuberc. 1958;77(4):623–43.

16. Lewinsohn DA, Zalwango S, Stein CM, Mayanja-Kizza H, Okwera A, Boom WH, et al. Whole blood interferon-gamma responses to mycobacterium tuberculosis antigens in young household contacts of persons with tuberculosis in Uganda. PLoS One. 2008;3(10):e3407.

17. Triasih R, Robertson C, Duke T, Graham SM. Risk of infection and disease with Mycobacterium tuberculosis among children identified through prospective community-based contact screening in Indonesia. Trop Med Int Health. 2015;20(6):737–43.

18. Amanullah F, Ashfaq M, Khowaja S, Parekh A, Salahuddin N, Lotia-Farrukh I, et al. High tuberculosis prevalence in children exposed at home to drug-resistant tuberculosis. Int J Tuberc Lung Dis. 2014;18(5):520–7.

19. Ma N, Zalwango S, Malone LL, Nsereko M, Wampande EM, Thiel BA, et al. Clinical and epidemiological characteristics of individuals resistant to M. tuberculosis infection in a longitudinal TB household contact study in Kampala, Uganda. BMC Infect Dis. 2014;14:352.

20. Rathi SK, Akhtar S, Rahbar MH, Azam SI. Prevalence and risk factors associated with tuberculin skin test positivity among household contacts of smear-positive pulnionary tuberculosis cases in Umerkot, Pakistan. Int J Tuberc Lung Dis. 2002;6(10):851–7.

21. Lienhardt C, Fielding K, Sillah J, Tunkara A, Donkor S, Manneh K, et al. Risk factors for tuberculosis infection in sub-Saharan Africa: a contact study in The Gambia. Am J Respir Crit Care Med. 2003;168(4):448–55.

22. Jones-Lopez EC, White LF, Kirenga B, Mumbowa F, Ssebidandi M, Moine S, et al. Cough Aerosol Cultures of Mycobacterium tuberculosis: Insights on TST / IGRA Discordance and Transmission Dynamics. PLoS One. 2015;10(9):e0138358.

23. Whalen CC, Zalwango S, Chiunda A, Malone L, Eisenach K, Joloba M, et al. Secondary attack rate of tuberculosis in urban households in Kampala, Uganda. PLoS One. 2011;6(2):e16137.

24. Kifai EJ, Bakari M. Mantoux skin test reactivity among household contacts of HIV-infected and HIV un-infected patients with sputum smear positive TB in Dar es Salaam, Tanzania. East Afr J Public Health. 2009;6(2):211–8.

25. Nunn P, Mungai M, Nyamwaya J, Gicheha C, Brindle RJ, Dunn DT, et al. The effect of human immunodeficiency virus type-1 on the infectiousness of tuberculosis. Tuber Lung Dis. 1994;75(1):25–32.

26. Espinal MA, Perez EN, Baez J, Henriquez L, Fernandez K, Lopez M, et al. Infectiousness of Mycobacterium tuberculosis in HIV-1-infected patients with tuberculosis: a prospective study. Lancet. 2000;355(9200):275–80.

27. Hesseling AC, Mandalakas AM, Kirchner HL, Chegou NN, Marais BJ, Stanley K, et al. Highly discordant T cell responses in individuals with recent exposure to household tuberculosis. Thorax. 2009;64(10):840–6.

28. Guwatudde D, Nakakeeto M, Jones-Lopez EC, Maganda A, Chiunda A, Mugerwa RD, et al. Tuberculosis in household contacts of infectious cases in Kampala, Uganda. Am J Epidemiol. 2003;158(9):887–98.

29. Lees AW, Allan GW, Smith J, Tyrrell WF. Pulmonary tuberculosis in contacts: a ten year survey. Dis Chest. 1961;40:516–21.


30. Ahmad Khan F, Verkuijl S, Parrish A, Chikwava F, Ntumy R, El-Sadr W, et al. Performance of symptom-based tuberculosis screening among people living with HIV: not as great as hoped. AIDS. 2014;28(10):1463–72.

31. Calnan M. Developing strategies for TB screening among HIV-infected and HIV-uninfected pregnant and postpartum women in Swaziland. 47th World Conference on Lung Health of the International Union Against Tuberculosis and Lung Disease; Liverpool, United Kingdom. 2016.

32. Hanifa Y, Fielding K, Chihota V, Ndlovu N, Karstaedt A, Adonis L, et al. Evaluation of WHO 4-symptom tool to rule out TB: Data from the XPHACTOR Study. Topics in Antiviral Medicine. 2015;23:372–3.

33. Kufa T, Mngomezulu V, Charalambous S, Hanifa Y, Fielding K, Grant AD, et al. Undiagnosed tuberculosis among HIV clinic attendees: association with antiretroviral therapy and implications for intensified case finding, isoniazid preventive therapy, and infection control. J Acquir Immune Defic Syndr. 2012;60(2):e22–8.

34. LaCourse SM, Cranmer LM, Matemo D, Kinuthia J, Richardson BA, John-Stewart G, et al. Tuberculosis Case Finding in HIV-Infected Pregnant Women in Kenya Reveals Poor Performance of Symptom Screening and Rapid Diagnostic Tests. J Acquir Immune Defic Syndr. 2016;71(2):219–27.

35. Nguyen DT, Bang ND, Hung NQ, Beasley RP, Hwang LY, Graviss EA. Yield of chest radiograph in tuberculosis screening for HIV-infected persons at a district-level HIV clinic. Int J Tuberc Lung Dis. 2016;20(2):211–7.

36. Rangaka MX, Wilkinson RJ, Glynn JR, Boulle A, van Cutsem G, Goliath R, et al. Effect of antiretroviral therapy on the diagnostic accuracy of symptom screening for intensified tuberculosis case finding in a South African HIV clinic. Clin Infect Dis. 2012;55(12):1698–706.

37. den Boon S, White NW, van Lill SW, Borgdorff MW, Verver S, Lombard CJ, et al. An evaluation of symptom and chest radiographic screening in tuberculosis prevalence surveys. Int J Tuberc Lung Dis. 2006;10(8):876–82.

38. Adetifa IM, Kendall L, Bashorun A, Linda C, Omoleke S, Jeffries D, et al. A tuberculosis nationwide prevalence survey in Gambia, 2012. Bull World Health Organ. 2016;94(6):433–41.

39. Ayles H, Schaap A, Nota A, Sismanidis C, Tembwe R, De Haas P, et al. Prevalence of tuberculosis, HIV and respiratory symptoms in two Zambian communities: implications for tuberculosis control in the era of HIV. PLoS One. 2009;4(5):e5602.

40. Corbett EL, Zezai A, Cheung YB, Bandason T, Dauya E, Munyati SS, et al. Provider-initiated symptom screening for tuberculosis in Zimbabwe: diagnostic value and the effect of HIV status. Bull World Health Organ. 2010;88(1):13–21.

41. van’t Hoog AH, Meme HK, Laserson KF, Agaya JA, Muchiri BG, Githui WA, et al. Screening strategies for tuberculosis prevalence surveys: the value of chest radiography and symptoms. PLoS One. 2012;7(7):e38691.

42. Datta M, Radhamani MP, Sadacharam K, Selvaraj R, Rao DL, Rao RS, et al. Survey for tuberculosis in a tribal population in North Arcot District. Int J Tuberc Lung Dis. 2001;5(3):240–9.

43. Gopi PG, Subramani R, Radhakrishna S, Kolappan C, Sadacharam K, Devi TS, et al. A baseline survey of the prevalence of tuberculosis in a community in south India at the commencement of a DOTS programme. Int J Tuberc Lung Dis. 2003;7(12):1154–62.


44. Kapata N, Chanda-Kapata P, Ngosa W, Metitiri M, Klinkenberg E, Kalisvaart N, et al. The Prevalence of Tuberculosis in Zambia: Results from the First National TB Prevalence Survey, 2013–2014. PLoS One. 2016;11(1):e0146392.

45. Ministry of Health. Report National TB Prevalence Survey, 2002 Cambodia. Phnom Penh, National Tuberculosis Control Programme, 2005.

46. Ministry of Health. Report on National TB Prevalence Survey 2009-2010. Nay Pyi Taw, Ministry of Health, Department of Health, 2012.

47. Senkoro M, Mfinanga S, Egwaga S, Mtandu R, Kamara DV, Basra D, et al. Prevalence of pulmonary tuberculosis in adult population of Tanzania: a national survey, 2012. Int J Tuberc Lung Dis. 2016;20(8):1014–21.

48. Kebede AH, Alebachew Z, Tsegaye F, Lemma E, Abebe A, Agonafir M, et al. The first population-based national tuberculosis prevalence survey in Ethiopia, 2010–2011. Int J Tuberc Lung Dis. 2014;18(6):635–9.

49. Law I, Sylavanh P, Bounmala S, Nzabintwali F, Paboriboune P, Iem V, et al. The first national tuberculosis prevalence survey of Lao PDR (2010–2011). Trop Med Int Health. 2015;20(9):1146–54.

50. Auguste P, Tsertsvadze A, Pink J, Court R, Seedat F, Gurung T, et al. Accurate diagnosis of latent tuberculosis in children, people who are immunocompromised or at risk from immunosuppression and recent arrivals from countries with a high incidence of tuberculosis: systematic review and economic evaluation. Health technology assessment (Winchester, England). 2016;20(38):1–678.


52. Mathad JS, Bhosale R, Balasubramanian U, Kanade S, Mave V, Suryavanshi N, et al. Quantitative IFN-gamma and IL-2 Response Associated with Latent Tuberculosis Test Discordance in HIV-infected Pregnant Women. Am J Respir Crit Care Med. 2016;193(12):1421–8.


54. Rangaka MX, Wilkinson RJ, Boulle A, Glynn JR, Fielding K, van Cutsem G, et al. Isoniazid plus antiretroviral therapy to prevent tuberculosis: a randomised double-blind, placebo-controlled trial. Lancet. 2014;384(9944):682–90.

55. Sharma SK, Vashishtha R, Chauhan LS, Sreenivas V, Seth D. Comparison of TST and IGRA in Diagnosis of Latent Tuberculosis Infection in a High TB-Burden Setting. PLoS One. 2017;12(1):e0169539.

56. Alsdurf H, Hill PC, Matteelli A, Getahun H, Menzies D. The cascade of care in diagnosis and treatment of latent tuberculosis infection: a systematic review and meta-analysis. Lancet Infect Dis. 2016;16(11):1269–78.

57. WHO. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection Recommendations for a public health approach – Second edition. 2016.

58. Spyridis NP, Spyridis PG, Gelesme A, Sypsa V, Valianatou M, Metsou F, et al. The effectiveness of a 9-month regimen of isoniazid alone versus 3- and 4-month regimens of isoniazid plus rifampin for treatment of latent tuberculosis infection in children: results of an 11-year randomized study. Clin Infect Dis. 2007;45(6):715–22.



60. van Zyl S, Marais BJ, Hesseling AC, Gie RP, Beyers N, Schaaf HS. Adherence to anti-tuberculosis chemoprophylaxis and treatment in children. Int J Tuberc Lung Dis. 2006;10(1):13–8.

61. Stagg HR, Zenner D, Harris RJ, Munoz L, Lipman MC, Abubakar I. Treatment of latent tuberculosis infection: a network meta-analysis. Ann Intern Med. 2014;161(6):419–28.

62. Shepardson D, MacKenzie WR. Update on cost-effectiveness of a 12-dose regimen for latent tuberculous infection at new rifapentine prices. Int J Tuberc Lung Dis. 2014;18(6):751.


64. Sterling TR, Scott NA, Miro JM, Calvet G, La Rosa A, Infante R, et al. Three months of weekly rifapentine and isoniazid for treatment of Mycobacterium tuberculosis infection in HIV-coinfected persons. AIDS. 2016;30(10):1607–15.

65. Sterling TR, Villarino ME, Borisov AS, Shang N, Gordin F, Bliven-Sizemore E, et al. Three Months of Rifapentine and Isoniazid for Latent Tuberculosis Infection. New England Journal of Medicine. 2011;365(23):2155–66.

66. Villarino ME, Scott NA, Weis SE, Weiner M, Conde MB, Jones B, et al. Treatment for preventing tuberculosis in children and adolescents: a randomized clinical trial of a 3-month, 12-dose regimen of a combination of rifapentine and isoniazid. JAMA Pediatr. 2015;169(3):247–55.

67. WHO. Global TB Report. 2016.

68. Schaaf HS, Garcia-Prats AJ, Hesseling AC, Seddon JA. Managing multidrug-resistant tuberculosis in children: review of recent developments. Curr Opin Infect Dis. 2014;27(3):211–9.

69. Bamrah S, Brostrom R, Dorina F, Setik L, Song R, Kawamura LM, et al. Treatment for LTBI in contacts of MDR-TB patients, Federated States of Micronesia, 2009–2012. Int J Tuberc Lung Dis. 2014;18(8):912–8.

70. Garcia-Prats AJ, Zimri K, Mramba Z, Schaaf HS, Hesseling AC. Children exposed to multidrug-resistant tuberculosis at a home-based day care centre: a contact investigation. Int J Tuberc Lung Dis. 2014;18(11):1292–8.


72. Fox GJ, Schaaf HS, Mandalakas A, Chiappini E, Zumla A, Marais BJ. Preventing the spread of multidrug-resistant tuberculosis and protecting contacts of infectious cases. Clin Microbiol Infect. 2017;23(3):147–53.

73. Targeted tuberculin testing and treatment of latent tuberculosis infection. American Thoracic Society. MMWR Recommendations and reports : Morbidity and mortality weekly report Recommendations and reports. 2000;49(Rr-6):1–51.

74. Cain KP, Nelson LJ, Cegielski JP. Global policies and practices for managing persons exposed to multidrug-resistant tuberculosis. Int J Tuberc Lung Dis. 2010;14(3):269–74.


75. WHO. Companion Handbook to the WHO Guidelines for the Programmatic Management of Drug-Resistant Tuberculosis. 2014.

76. WHO. Active TB drug-safety monitoring and management (aDSM). Framework for implementation. 2015.

77. Kritski AL, Marques MJ, Rabahi MF, Vieira MA, Werneck-Barroso E, Carvalho CE, et al. Transmission of tuberculosis to close contacts of patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med. 1996;153(1):331–5.



Documents

Latent tuberculosis infection latent TB -201… · Diane Havlir declared that she received research support from National Institutes of Health for research on TB. Marieke van der