The n e w e ng l a n d j o u r na l of m e dic i n e

Clinical Practice

Caren G. Solomon, M.D., M.P.H., Editor

Latent Tuberculosis Infection

Maunank Shah, M.D., Ph.D., and Susan E. Dorman, M.D.​​

This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence
­supporting various strategies is then presented, followed by a review of formal guidelines, when they exist.
The article ends with the authors’ clinical recommendations.

A 36-year-old man comes to establish care with a primary care clinician. He has no From the Johns Hopkins University
chronic medical conditions and takes no medications. He was born and lived in School of Medicine and the Baltimore
City Health Department — both in Balti-
­India until 1 year ago, when he came to the United States. He reports never having more (M.S.); and the Medical University
had tuberculosis or tuberculosis testing, although when the patient was a child, of South Carolina, Charleston, and the
his father was treated for tuberculosis disease. The patient received bacille South Carolina Department of Health
and Environmental Control, Columbia
Calmette–Guérin vaccine as an infant. He reports no cough, fever, night sweats, or (S.E.D.). Dr. Dorman can be contacted at
weight loss. Should testing for Mycobacterium tuberculosis infection be performed ­dorman@​­musc​.­edu or at the Medical
and, if so, how? If testing shows evidence of M. tuberculosis infection, how should University of South Carolina, 135 Rutledge
Ave., Rm. 1207, Charleston, SC 29425.
this patient be treated?
N Engl J Med 2021;385:2271-80.
DOI: 10.1056/NEJMcp2108501
The Cl inic a l Probl em Copyright © 2021 Massachusetts Medical Society.

O
ne fourth of the world’s population is infected with Mycobac- CME
at NEJM.org
terium tuberculosis, the causative agent of tuberculosis.1 From a clinical-
management perspective, two states of M. tuberculosis infection are recog-
nized — latent tuberculosis infection (LTBI) and active tuberculosis disease — although
infection and immunologic control exist across a spectrum. LTBI is a state of
persistent immune response to stimulation by M. tuberculosis antigens without
evidence of clinically manifested active tuberculosis and with bacillary replica-
tion absent or below some undefined threshold as a result of immunologic
An audio version
control. Most persons with LTBI never become sick with tuberculosis; however,
of this article
approximately 5 to 15% have progression to tuberculosis disease.2-5 Tuberculosis is available at
disease results from bacterial replication and includes tissue damage and in- NEJM.org
flammation, which lead to clinical manifestations. Common signs and symp-
toms include fever, sweats, fatigue, and weight loss, as well as cough if the lungs
are affected. Tuberculosis disease can involve any tissue and organ system, with
diverse manifestations.
Prevention of progression from LTBI to tuberculosis disease is an important
individual and public health goal because persons with LTBI are a reservoir for
incident tuberculosis disease. Two principles form the basis for guidance about
whom to test for LTBI and whom to treat for LTBI. First, infection with M. tuber-
culosis requires previous exposure to someone with tuberculosis disease, and there-
fore asking about known exposure and epidemiologic risk factors for exposure is
important. Second, for a person infected with M. tuberculosis, the risk of progres-
sion to tuberculosis disease is determined by their immune competence and the
time since initial infection; therefore, eliciting biologic risk factors for progression
is important.

n engl j med 385;24  nejm.org  December 9, 2021 2271

The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Key Clinical Points

Latent Tuberculosis Infection

• Prevention of progression from latent tuberculosis infection (LTBI) to tuberculosis disease is an
important individual and public health goal.
• Adults and children should be screened for risk factors for Mycobacterium tuberculosis exposure and
for risk factors for progression to tuberculosis disease. Persons who screen positive should be tested
for M. tuberculosis infection, preferably with the use of an interferon-γ release assay. Persons who test
positive for M. tuberculosis infection should be assessed for tuberculosis disease.
• Persons with LTBI who are at increased risk for progression to tuberculosis disease generally should
be treated for LTBI and followed until treatment is completed.
• Preferred LTBI treatment regimens include 3 months of once-weekly rifapentine plus isoniazid,
4 months of once-daily rifampin, or 3 months of once-daily isoniazid plus rifampin. Isoniazid
administered once daily for 6 or 9 months is an alternative.
• Risk factors for hepatotoxic effects and drug–drug interactions should be considered when the LTBI
treatment regimen is selected.

S t r ategie s a nd E v idence Several other medical conditions are associ-

Identification of Persons Who Should
Be Tested for LTBI
All adults and children should have an assess-
ment of risk factors for M. tuberculosis exposure
and risk factors for progression from LTBI to
tuberculosis disease at least once as a compo-
nent of routine primary care (Fig. 1).6 Exposure
risk is driven by tuberculosis epidemiology and
exposure history. In 2020, the United States had
a tuberculosis incidence of 2.2 per 100,000 per-
sons, but most countries in Africa, Asia, Eastern
Europe, Latin America, and the Pacific Islands
have substantially higher incidence.7,8 Persons
who were born, lived, or had prolonged travel
(>1 month) in these higher-incidence settings
should undergo testing for LTBI regardless of
duration of residence in the United States (Ta-
ble 1).9 Within lower-incidence countries, there
are specific settings (e.g., correctional facilities
and homeless shelters) with higher tuberculosis
incidence that warrant testing for LTBI; local
health departments can provide guidance about
epidemiology and local regulatory requirements
for LTBI testing.6,9 Close contact with a person
with infectious tuberculosis disease is a risk fac-
tor for M. tuberculosis infection, with the risk of
progression to tuberculosis disease being high-
est in the first 2 years after infection. Persons
with immunosuppressive conditions, including
human immunodeficiency virus (HIV) infection
and iatrogenic immunosuppression, typically have
substantially increased risk of progression from
LTBI to tuberculosis disease, such that routine
testing for LTBI is indicated.3,6
ated with a modestly increased risk of progres-
sion from LTBI to tuberculosis disease but are
not considered to be independent indications for
LTBI testing (Table 1).3,6 Testing for M. tuberculosis
infection generally is not warranted in asymp-
tomatic persons without epidemiologic risk of
infection or biologic risk factors for progression
to tuberculosis disease, because the positive pre-
dictive value of available tests is low when the
pretest probability of infection is low.6,10
Diagnosis and Evaluation
Tests to Detect M. tuberculosis Infection
There are no direct microbiologic tests for diag-
nosis of LTBI. Instead, interferon-γ release as-
says (IGRAs) and tuberculin skin testing provide
indirect assessment of infection through detec-
tion of cell-mediated immune responses to stimu-
lation by M. tuberculosis antigens. These tests
cannot discriminate LTBI from tuberculosis dis-
ease, and they cannot be used to monitor treat-
ment efficacy because immunologic responses
that are detected by the tests are long-lasting
even after effective treatment. Current U.S.
guidelines give preference to the use of IGRAs
in adults and older children, but tuberculin skin
tests are considered to be acceptable.6,10
The tuberculin skin test and IGRAs differ
with respect to practical aspects, readout, and
accuracy (Fig. 2).11 IGRAs are in vitro assays that
require phlebotomy and have objective results
but rely on laboratory expertise, equipment, and
reagents. In vivo tuberculin skin testing requires
two clinical visits plus administration and inter-
pretation by a trained provider. IGRAs are more

2272 n engl j med 385;24  nejm.org  December 9, 2021

The New England Journal of Medicine

Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Clinical Pr actice

Screening (through interview, questionnaire, and medical history review) to assess for
risk factors for infection with M. tuberculosis and risk factors for progression
to active tuberculosis if infection with M. tuberculosis is present

Any one or more high-priority No high-priority

risk factors present risk factor

Perform testing to evaluate for Testing generally not indicated;

M. tuberculosis infection periodically rescreen
(interferon-γ release assays to assess for emergence
preferred over tuberculin skin test) of risk factors

Testing result positive for Testing result negative for

M. tuberculosis infection M. tuberculosis infection

Evaluate for active tuberculosis disease No additional tuberculosis testing indicated

Signs and symptoms for asymptomatic, nonimmunocompromised
Chest radiography persons; if host is at high risk for progression
and has high epidemiologic risk, consider
repeat testing or testing with another assay,
interpreting any positive test as evidence
supporting M. tuberculosis infection

Signs, symptoms, or No evidence of active

chest radiograph suggestive of tuberculosis disease
active tuberculosis disease

Perform additional tests to evaluate No evidence of active Assess potential benefits vs. potential risks of
for active tuberculosis disease tuberculosis disease LTBI treatment
Biologic risk factors for progression from
Microbiologic or LTBI to active tuberculosis disease
clinical diagnosis of active Drug–drug interaction
tuberculosis disease Risk of hepatotoxic effects
Pregnancy testing

Treat for active tuberculosis disease

Potential benefits of Potential risks of
LTBI treatment LTBI treatment
outweigh potential risks outweigh potential benefits
(applies to most persons) (less common)

Educate patient about tuberculosis signs

Treat for LTBI
and symptoms
Document evidence of LTBI in medical record
Reevaluate risk–benefit ratio over time as
patient’s medical conditions change

Figure 1. Algorithm for Deciding Whom to Test and Whom to Treat for Latent Tuberculosis Infection (LTBI).
This algorithm is intended for use in asymptomatic persons and is not intended for use in the evaluation of persons who present for
medical care with signs or symptoms of active tuberculosis disease. High-priority risk factors are household or other close contact with
someone with infectious active tuberculosis disease; birth, residence, or prolonged travel (>1 mo) in a setting in which tuberculosis dis-
ease is common; other circumstances based on local epidemiology (e.g., corrections facilities and homeless shelters); and immunosup-
pression. Biologic risk factors for progression of LTBI to active tuberculosis disease are immunosuppression; diabetes mellitus; chronic
kidney disease; leukemia or lymphoma; cancer of the head or neck; chronic malabsorption, gastrectomy, or intestinal bypass; a body-
mass index (the weight in kilograms divided by the square of the height in meters) of 20 or less; silicosis; current or former smoking sta-
tus; or an age of 5 years or younger.

n engl j med 385;24  nejm.org  December 9, 2021 2273

The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Risk-Based Approach for Latent Tuberculosis Infection (LTBI) Testing and Treatment.*

Findings Based on History Recommended Action

Any of the following high-priority risk factors†:
Household or other close contact with someone with infectious active tuberculosis
disease‡
Birth, residence, or prolonged travel (>1 mo) in a setting in which tuberculosis dis-
ease is common (most countries in Africa, Asia, Eastern Europe, Latin America,
and the Pacific Islands)§
Other circumstances based on local epidemiology (e.g., corrections facilities and
homeless shelters)§
Immunosuppression (HIV infection or current or planned immunosuppression with
the use of a TNF-α antagonist, glucocorticoids equivalent to prednisone at a dose
of 2 mg per kilogram of body weight per day or 15 mg per day for ≥1 mo, or other
immunosuppressing medications)‡
Any of the following biologic factors that confer generally modest increased risk of
progression to active tuberculosis disease if M. tuberculosis infection is present§:
diabetes mellitus; chronic kidney disease; leukemia or lymphoma; cancer of the
head or neck; chronic malabsorption, gastrectomy, or intestinal bypass; a body-
mass index (the weight in kilograms divided by the square of the height in meters)
of ≤20; silicosis; current or former smoking status; or an age of ≤5 yr¶
No epidemiologic risk factors for tuberculosis exposure and no host risk factors for
progression to active tuberculosis disease if M. tuberculosis infection is present
Testing for M. tuberculosis infection should be
performed. If LTBI is present, LTBI treatment
is recommended unless there are medical con­
traindications or clinically significant drug–drug
interactions that cannot be mitigated.
In the absence of epidemiologic risk factors (as
noted above) for tuberculosis exposure, testing
for M. tuberculosis infection is generally not rec-
ommended. However, if LTBI is present, LTBI
treatment is recommended unless there are medi-
cal contraindications or clinically significant drug–
drug interactions that cannot be mitigated.
Testing for M. tuberculosis infection is not recom-
mended.

* This table is based on recommendations from the National Society of Tuberculosis Clinicians.6 HIV denotes human immunodeficiency virus,
and TNF-α tumor necrosis factor α.
† Tuberculin skin test (TST) results are interpreted at the listed positivity thresholds on the basis of epidemiologic or host risk and to improve
test positive and negative predictive values. Interferon-γ release assays currently have absolute thresholds for positivity without incorpora-
tion of host or epidemiologic risk.
‡ The TST positivity threshold is 5 mm or more.
§ The TST positivity threshold is 10 mm or more.
¶ Children 5 years of age or younger are at high risk for rapid progression to active tuberculosis disease after infection, and severe forms of
tuberculosis (including disseminated and central nervous system tuberculosis) are more likely to develop in such children than in older
children or in adults.

specific than tuberculin skin testing for the de-
tection of M. tuberculosis infection. Vaccination
against bacille Calmette–Guérin (BCG) can cause
a positive tuberculin skin test owing to shared
antigenic components with purified protein de-
rivative, although the effect on the tuberculin
skin test of BCG received in infancy is minimal,
especially 10 years or more after vaccination.12
IGRA results are unaffected by previous BCG
vaccination, because stimulatory antigens that
are used in these assays are absent from BCG
vaccines. In settings of low tuberculosis preva-
lence, the specificity of IGRAs is more than
95%; the specificity of tuberculin skin tests is
also more than 95% in populations without BCG
vaccination but is lower in populations in which
BCG vaccines are administered.13,14 Nevertheless,
IGRAs can yield false positive results, particu-
larly when used in low-risk populations.15
Estimated sensitivities of tuberculin skin tests
and IGRAs have relied on the assessment of per-
formance in patients with tuberculosis disease
as a reference standard and have varied widely.
The T-SPOT.TB and QuantiFERON-TB Gold Plus
IGRAs have sensitivities of approximately 90%,
whereas that of tuberculin skin tests is approxi-
mately 80%.13,14,16 Host immunocompromise, in-
cluding that arising from tuberculosis disease
itself, can be associated with false negative skin
tests and IGRAs.17,18 For IGRAs and tuberculin skin
testing, there are positive associations between
the risk of progression to tuberculosis disease
and test quantitative results, but this popula-
tion-level observation is not readily translatable
to the individual patient because immunocom-
promise can cause low quantitative results in a
person at high risk for tuberculosis disease.19,20
Evaluation of Persons with Positive Test Results
A diagnosis of LTBI requires the ruling out of
tuberculosis disease by assessment for signs and
symptoms of tuberculosis disease plus chest ra-
diography. Findings suggestive of tuberculosis
disease should trigger individualized evaluation
that usually includes obtaining sputum speci-
mens for mycobacterial culture and performing

2274 n engl j med 385;24  nejm.org  December 9, 2021

The New England Journal of Medicine

Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Clinical Pr actice

smear microscopy and nucleic acid amplification regimen had similar efficacy to the 9-month
testing and that may include further imaging, regimen (0.10 and 0.11 cases of active tubercu-
bronchoscopy, and biopsy of affected tissues. losis disease per 100 person-years, respectively)
and was associated with a higher completion
Antimicrobial Treatment for LTBI rate (78.8% and 63.2%, respectively; P<0.001),
Identification of Appropriate Candidates for Treatment fewer overall adverse events, and fewer hepa-
Randomized, controlled trials have shown that totoxic effects (0.3% and 1.8%, respectively;
antimicrobial treatment of LTBI is effective in P<0.001).31 Retrospective observational studies
preventing progression to tuberculosis disease involving patients at public health clinics in
overall and in children and persons living with North America had similar findings.32-34 A regi-
HIV and reducing mortality among adults with men of 3 months of once-daily isoniazid plus
HIV infection.21-25 Decisions about whether to rifampin has been shown to have an efficacy
recommend LTBI treatment are based on risks of and safety that are similar to those of daily iso-
adverse effects from treatment (discussed below) niazid administered for 6 to 12 months.35
weighed against the risk of development of tu- Rifampin and rifapentine-containing LTBI
berculosis disease. Persons with LTBI who are treatment regimens are less commonly associ-
at increased risk for progression to tuberculosis ated with hepatotoxic effects than is isonia-
disease (Table 1) generally should be offered zid.26,31,34,36 However, rifampin and rifapentine
LTBI treatment to prevent illness and death asso- induce cytochrome P450 enzymes and transport-
ciated with tuberculosis disease.6 All persons with ers, causing reductions in the plasma concentra-
LTBI should undergo testing for HIV infection. tions of certain drugs (Table 2). Rifamycins can
cause neutropenia or thrombocytopenia as well
Rifamycin-Containing Regimens (Preferred) as a flulike systemic drug reaction that rarely
Rifamycin-based treatments are now the pre- includes hypotension and syncope41 (Table 2).
ferred approach for LTBI treatment because they
have similar or better efficacy and higher com- Isoniazid Monotherapy (Alternative)
pletion rates than isoniazid monotherapy. The Isoniazid monotherapy has long been used for
randomized, controlled PREVENT TB trial com- LTBI treatment. However, given its longer treat-
pared 3 months of once-weekly rifapentine plus ment duration, lower completion rates, and high-
isoniazid, administered by directly observed ther- er rates of hepatotoxic effects as compared with
apy, with 9 months of participant-administered rifamycin-based regimens, isoniazid monother-
daily isoniazid in 7731 persons at high risk for apy is no longer considered in U.S. guidelines to
tuberculosis disease. In the trial, a 3-month course be a preferred LTBI treatment.6 Isoniazid de-
of rifapentine and isoniazid was noninferior to creases the risk of tuberculosis disease among
a 9-month course of isoniazid with respect to persons with a positive tuberculin skin test, in-
the cumulative rate of tuberculosis disease (0.19% cluding adults and children without HIV infec-
vs. 0.43%) and was associated with a higher rate tion and adults with HIV infection.21-23 A pre–
of treatment completion (82.1% vs. 69.0%, HIV era trial that involved approximately 28,000
P<0.001) and a lower rate of drug-related hepa- persons at high risk for progression to tuber-
totoxic effects (0.4% vs. 2.7%, P<0.001).26 Effi- culosis disease showed that treatment with 12
cacy similar to that of longer isoniazid regi- months or 6 months of isoniazid was more ef-
mens, a high completion rate, and a good safety fective than treatment for 3 months (75%, 65%,
profile were confirmed in studies involving chil- and 21% lower incidence of tuberculosis disease,
dren 2 years of age or older and in studies in- respectively, than with placebo), thereby estab-
volving adults with HIV infection.27-29 A phase 4 lishing 6 to 12 months as the target duration.39,40
randomized trial showed a similarly high rate of The effectiveness of isoniazid monotherapy is
completion of a participant-administered 3-month diminished by completion rates of 50% or less
course of rifapentine and isoniazid in the United in clinical practice.42 Isoniazid can cause clini-
States and elsewhere.30 cally significant hepatotoxic effects, which oc-
A multinational, randomized, controlled trial curred in 2 to 3% of persons receiving isoniazid
that involved more than 6000 adults compared LTBI treatment in clinical trials26,31,36 (Table 2).
4 months of once-daily rifampin with 9 months In a retrospective cohort of all persons starting
of isoniazid for LTBI treatment.31 The 4-month isoniazid LTBI treatment in Quebec, Canada,

n engl j med 385;24  nejm.org  December 9, 2021 2275

The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Tuberculin Skin Test T-SPOT.TB IGRA QuantiFERON-TB Gold Plus IGRA

Phlebotomy Phlebotomy
Intradermal inoculation
of PPD into volar
aspect of forearm

Whole blood
Whole blood
PBMCs Add blood to
tubes precoated
Centrifugation with nil control, M.
tuberculosis antigens
(tubes TB1 and
Wait 48–72 hr. TB2), and mitogen
Nil TB1 TB2 Mitogen positive control.
Plate 250,000 cells into
each of four wells per Incubate 16–24 hr. Add aliquot
patient. of stimulated plasma to wells
of ELISA plate that contains
interferon-γ antibodies.
Add nil control, M.
tuberculosis antigens
(×2), and mitogen
positive control.
Incubate for 16–20 hr.

0 1 2 3 4 5 6 7
Interferon-γ antibodies
Tuberculin Skin Test Ruler capture interferon-γ as it A secondary enzyme-
is released from cells. A linked antibody is added
secondary enzyme-labeled and binds to interferon-γ.
Measure and record antibody is added and binds A detection reagent is
diameter of induration. to captured interferon-γ. A added, and absorbance at
detection reagent is added, resulting in spots that 450 nm is measured. Con-
are a footprint of the location where the inter- centration of interferon-γ
feron-γ was released by a cell. Spots are counted. is calculated on the basis
of a standard curve.

In vivo assay In vitro enzyme-linked immunosorbent In vitro ELISA

spot assay

No instrumentation Laboratory instrumentation required; assay results can be

affected by manufacturing, preanalytic, and analytic factors11

Two patient encounters required to One patient encounter sufficient to obtain result
obtain result

Interpretation subjective; positivity Interpretation less subjective to not Interpretation not subjective; positivity
thresholds are risk-stratified subjective; positivity thresholds fixed thresholds fixed

Cross-reacts with BCG (PPD consists Does not cross-react with BCG on the basis of
of many components) selection of stimulation antigens

A previous tuberculin skin test can boost A previous IGRA does not boost a subsequent
a subsequent such test or IGRA tuberculin skin test or IGRA

between 2003 and 2007, a total of 15 of 9684 Regimen Selection

(0.2%) had severe hepatotoxic effects, including
1 adult who died and 2 who received liver trans-
plants.34 Isoniazid can cause peripheral neuropa-
thy, which can be mitigated by administration of
pyridoxine.
Regimen selection is influenced by coexisting
conditions and drug–drug interactions (Table 2).
The presence of risk factors for hepatotoxic ef-
fects (e.g., previous viral hepatitis, known liver
disease, regular alcohol use, injection drug use,

2276 n engl j med 385;24 nejm.org December 9, 2021

The New England Journal of Medicine

Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Clinical Pr actice
Figure 2 (facing page). Key Features of Tests
for the Detection of Infection with M. tuberculosis.
The T-SPOT.TB interferon-γ release assay (IGRA) includes
a result category of “borderline,” which reflects results
that approach but do not reach the threshold for posi-
tivity. In such cases, clinicians can repeat the test, per-
form a different test, or interpret the test on the basis
of pretest probability (i.e., as negative in a person with
a low likelihood of infection and low risk of progression
to tuberculosis disease and as positive if the likelihood of
infection or progression is high). The QuantiFERON-TB
Gold Plus IGRA includes a result category of “indetermi-
nate.” Indeterminate results represent either a low inter­
feron-γ concentration from the mitogen positive control
tube or a high interferon-γ concentration from the nil
negative control tube. Indeterminate QuantiFERON-TB
Gold Plus results provide no information about the M. tu-
berculosis infection status of the patient; in such cases,
clinicians should repeat the test or use a different test.
In some persons who are infected with M. tuberculosis,
a delayed-type hypersensitivity response to the tuberculin
skin test can wane, especially after many years, resulting
in a negative tuberculin skin test. However, the intrader-
mal inoculation of purified protein derivative (PPD) can
itself stimulate the ability to mount a delayed-type hyper-
sensitivity reaction, thereby causing a positive reaction
(a “boost”) on subsequent tuberculin skin tests. This new-
ly positive (“boosted”) reaction could be incorrectly inter-
preted as evidence of new M. tuberculosis infection. BCG
denotes bacille Calmette–Guérin vaccine strain, ELISA
enzyme-linked immunosorbent assay, and PBMCs periph-
eral-blood mononuclear cells.
pregnancy or less than 3 months post partum,
medications with potential hepatotoxic effects,
and older age) should be determined by medical
history; if present, baseline laboratory testing
for measurements of serum aspartate amino-
transferase (AST) and alanine aminotransferase
(ALT) should be performed. Daily rifampin or
once-weekly isoniazid plus rifapentine is gener-
ally preferred over isoniazid monotherapy in
persons with underlying risks for hepatotoxic
effects. Pregnancy testing should be performed
in persons of childbearing potential. The regi-
men of weekly rifapentine plus isoniazid has not
been adequately studied in pregnancy and is not
currently recommended6,37; detailed guidance is
provided in the Supplementary Appendix, which
is available with the full text of this article at
NEJM.org. Concomitant medications should be
reviewed for potential drug–drug interactions
with antimicrobial agents administered for LTBI
treatment (Table 2). A baseline complete blood
count is recommended for persons with coexist-
ing conditions.
n engl j med 385;24  nejm.org  December 9, 2021
The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
Monitoring during Treatment
Monitoring should be performed at least month-
ly to assess for adverse effects, adherence, and
signs or symptoms suggestive of active tubercu-
losis disease. Assessment of AST and ALT levels
and a complete blood count should be per-
formed if there are signs or symptoms of poten-
tial hepatotoxic effects, hypersensitivity reaction,
or easy bruising or bleeding. These tests should
be performed periodically in asymptomatic per-
sons with previous abnormal aminotransferase
levels, heavy alcohol use, or other risks for hepa-
totoxic effects.
A r e a s of Uncer ta in t y
Strategies are needed to better identify, and
thereby prioritize for preventive treatment, the
subgroup of M. tuberculosis–infected persons who
will have progression to tuberculosis disease.
Although the status of M. tuberculosis infection is
often viewed as binary (i.e., LTBI vs. tuberculosis
disease), this is inaccurate. Infection exists within
a continuous spectrum of bacterial metabolic
activity and immunologic responses and includes
intermediate states of “incipient” and “subclinical”
tuberculosis (Fig. S1 in the Supplementary Ap-
pendix).43,44 Biomarkers are needed to distinguish
persons at high risk for progression or who are
already having progression from those in whom
any persisting M. tuberculosis has been eliminated
or is being effectively controlled; limited data sup-
port associations between certain host transcrip-
tional signatures and the risk of progression.45
More data are needed on the efficacy and
safety of ultrashort LTBI treatments. The ran-
domized, controlled BRIEF TB/A5279 (Brief
Rifapentine–Isoniazid Efficacy for TB [Tubercu-
losis] Prevention/A5279) trial compared 1 month
of daily rifapentine plus isoniazid with 9 months
of isoniazid in 3000 participants with HIV who
were receiving antiretroviral therapy 36; the 1-month
regimen was noninferior to 9 months of isonia-
zid for the primary composite end point of tu-
berculosis diagnosis, death from tuberculosis, or
death from unknown cause (incidence rate, 0.65
per 100 person-years and 0.67 per 100 person-years,
respectively)36 and was associated with a signifi-
cantly higher completion rate (97% and 90%,
respectively). The 1-month regimen is included
as an alternative choice in World Health Organi-
zation (WHO) guidelines but not in current U.S.
2277
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Dose, Frequency, and Prescribing Information for Recommended Regimens for Treatment of LTBI.*

Dose for Adults and Children

Priority and Regimen†
Preferred
Isoniazid plus rifapentine
once weekly for 3 mo
(12 doses)§
Rifampin once daily for 4 mo
Isoniazid plus rifampin once
daily for 3 mo
Alternative
Isoniazid once daily for
6 mo‖**
Isoniazid once daily for
9 mo‖††
≥12 Yr of Age37,38
Isoniazid: 15 mg/kg/dose rounded up
to nearest 50 or 100 mg; maximum
dose, 900 mg
Rifapentine: 750 mg per dose if weight
is 32.1–49.9 kg; 900 mg per dose if
weight is ≥50 kg; maximum dose,
900 mg
10 mg/kg/day; maximum daily dose,
600 mg
Isoniazid: 5 mg/kg/day; maximum daily
dose, 300 mg
Rifampin: 10 mg/kg/day; maximum
daily dose, 600 mg
5 mg/kg/day; maximum daily dose,
300 mg
Same as above
Additional Prescribing Information‡
Administration: taking with high-fat foods increases rifapentine
­absorption and is recommended. Avoid concomitant aluminum-
containing antacids and foods with high monoamine content.
Adverse reactions: possible hypersensitivity reaction (3.8%), rash
(0.8%), hepatotoxic effects (0.4%).26 Hypersensitivity reactions
can include hypotension, bronchospasm, angioedema, conjunc-
tivitis, and urticaria.
Drug–drug interactions: rifapentine causes reductions in plasma
concentrations of certain drugs, including warfarin, apixaban,
rivaroxaban, dabigatran, hormonal contraceptives, levothyrox-
ine, methadone, and many HIV antiretroviral drugs. The effect
of once-weekly rifapentine appears to be less than that of daily
rifampin, but data are limited. For interactions with isoniazid,
see below.
Administration: taking on an empty stomach is preferable, if side
effects are acceptable.
Adverse reactions: hepatotoxic effects (0.3%), rash or other allergy
(0.2%), hematologic toxic effects (0.2%), unacceptable GI ad-
verse events (0.1%).31
Drug–drug interactions: as for rifapentine, above.
Administration: taking on an empty stomach is preferable, if side
effects are acceptable. Avoid concomitant aluminum-containing
antacids and foods with high monoamine content.
Adverse reactions: limited published data; hepatotoxic effects
(1–6%), rash (1–8%), unacceptable GI adverse events (0–6%).35
Isoniazid can cause peripheral neuropathy that can be mitigated
by pyridoxine (25–50 mg/day).¶
Drug–drug interactions: as for rifapentine (above) and isoniazid
(below).
Administration: taking on an empty stomach is preferable, if side
effects are acceptable. Avoid concomitant aluminum-containing
antacids and foods with high monoamine content.
Adverse reactions: hepatotoxic effects (2–3%), rash (0.6%), possible
hypersensitivity (0.5%).26,31,36 Isoniazid can cause peripheral neu-
ropathy that can be mitigated by pyridoxine (25–50 mg/day).¶
Drug–drug interactions: Isoniazid can increase the serum concen-
trations of carbamazepine, phenytoin, warfarin, disulfiram, and
others. Isoniazid can decrease the serum concentrations of itra-
conazole and ketoconazole.
Same as above

* GI denotes gastrointestinal.

† With respect to priority, “preferred” indicates excellent side-effect profile and efficacy, shorter treatment duration, and higher completion
rates than longer regimens and therefore higher effectiveness; “alternative” indicates good efficacy but lower completion rates than shorter
regimens and therefore lower effectiveness. The guidelines for regimens do not apply in circumstances in which there is evidence that the
infecting strain of M. tuberculosis is resistant to both isoniazid and rifampin.
‡ An online resources for up-to-date information about drug–drug interactions is the Medscape Drug Interaction Checker: https://reference​
.­medscape​.­com/​­drug​-­interactionchecker. Information about possible interactions between tuberculosis drugs and HIV drugs is available
at https://clinicalinfo​.­hiv​.­gov/​­en/​­guidelines/​­adult​-­and​-­adolescent​-­arv.
§ A regimen of once-weekly isoniazid plus rifapentine is not recommended for use in pregnant persons or those who anticipate becoming
pregnant during the treatment period because its safety in these populations has not been adequately studied.
¶ Pyridoxine is recommended for persons with preexisting neuropathy or neuropathy risk factors such as diabetes or HIV infection.
‖ Isoniazid may also be administered twice weekly by directly observed therapy for 6 or 9 months, at a dose of 15 mg per kilogram per dose
for adults.
** Once-daily isoniazid is strongly recommended for HIV-negative adults who are unable to take a preferred regimen. For persons living with
HIV infection, either 6 or 9 months of isoniazid can be used.
†† The recommendation for a 9-month regimen emerged after a reanalysis of data showed that increased protection was obtained with 9 to
10 months of treatment; clinical-trial data are lacking that directly compare 9 months of isoniazid with other durations.39,40

2278 n engl j med 385;24  nejm.org  December 9, 2021

The New England Journal of Medicine

Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Clinical Pr actice

guidelines, and its efficacy and safety have not
been established beyond the studied population
of adults and adolescents living with HIV mostly
in settings with a high incidence of tuberculo-
sis.46 The roles of new drugs for LTBI treatment
(ClinicalTrials.gov number, NCT03568383) and of
vaccines for preventing infection and progression
(NCT04453293 and NCT03512249) are being in-
vestigated.47,48 Inclusion of children and pregnant
persons in LTBI treatment trials is essential for
informing safety and efficacy in these populations
and accelerating their access to new regimens.49
Guidel ine s
Guidelines for the evaluation and management
of LTBI have been issued by the WHO and re-
gional entities.45 Recommendations differ some-
what across guidelines, reflecting regional dif-
ferences in tuberculosis epidemiology and health
systems. Recommendations in this article are
concordant with Centers for Disease Control and
Prevention guidelines.6,10,37,38
C onclusions a nd
R ec om mendat ions
The asymptomatic patient in the vignette has
epidemiologic risk factors for M. tuberculosis in-
fection, and testing for M. tuberculosis infection is
indicated. Use of an IGRA is preferred over tu-
berculin skin testing and has specificity advan-
tages in this person who received a BCG vaccine.
If testing indicates M. tuberculosis infection, chest
radiography and clinical assessment should be
performed to evaluate for tuberculosis disease,
and he should undergo testing for HIV infection.
Clinical assessment should elucidate coexisting
conditions that increase the risk of progression
from LTBI to tuberculosis disease, if this infor-
mation has not already been gathered. We would
assess AST and ALT levels and perform a com-
plete blood count before initiating therapy, al-
though these studies are not strictly recommend-
ed in current guidelines for persons without
coexisting conditions or risk factors for hepato-
toxic effects. If there is no evidence of tubercu-
losis disease and aminotransferase levels are nor-
mal, then LTBI treatment should be initiated with
one of the preferred rifamycin-based regimens,
given their similar or greater efficacy and higher
completion rates as compared with isoniazid
monotherapy. Our practice would be to recom-
mend 3 months of once-weekly rifapentine plus
isoniazid; an alternative would be 4 months of
daily rifampin. Monthly clinical follow-up is war-
ranted to assess for medication adverse effects
and adherence and for any signs or symptoms
that could indicate progression to tuberculosis
disease. If pretreatment aminotransferase results
were normal for a person taking a rifamycin-
containing preferred regimen, our practice would
be to check aminotransferase levels and perform
a complete blood count after 1 month of treat-
ment and, if findings were normal, thereafter
monitor clinically without laboratory testing un-
less there were new symptoms of an adverse
drug reaction.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.

References
1. Cohen A, Mathiasen VD, Schön T,
Wejse C. The global prevalence of latent
tuberculosis: a systematic review and meta-
analysis. Eur Respir J 2019;​54:​1900655.
2. Comstock GW, Livesay VT, Woolpert
SF. The prognosis of a positive tuberculin
reaction in childhood and adolescence.
Am J Epidemiol 1974;​99:​131-8.
3. Campbell JR, Winters N, Menzies D.
Absolute risk of tuberculosis among un-
treated populations with a positive tuber-
culin skin test or interferon-gamma re-
lease assay result: systematic review and
meta-analysis. BMJ 2020;​368:​m549.
4. Getahun H, Matteelli A, Chaisson RE,
Raviglione M. Latent Mycobacterium tuber-
culosis infection. N Engl J Med 2015;​372:​
2127-35.
5. Horsburgh CR Jr, Rubin EJ. Latent tu-
berculosis infection in the United States.
N Engl J Med 2011;​364:​1441-8.
6. National Society of Tuberculosis Cli-
nicians. Testing and treatment of latent
tuberculosis infection in the United States:​
clinical recommendations. Smyrna, GA:​
National Tuberculosis Controllers Asso-
ciation, February 2021.
7. Deutsch-Feldman M, Pratt RH, Price
SF, Tsang CA, Self JL. Tuberculosis —
United States, 2020. MMWR Morb Mortal
Wkly Rep 2021;​70:​409-14.
8. Global tuberculosis report 2021. Ge-
neva:​World Health Organization, 2021.
9. US Preventive Services Task Force.
Screening for latent tuberculosis infec-
tion in adults: US Preventive Services Task
Force recommendation statement. JAMA
2016;​316:​962-9.
10. Lewinsohn DM, Leonard MK, LoBue
PA, et al. Official American Thoracic Society/
Infectious Diseases Society of America/
Centers for Disease Control and Preven-
tion clinical practice guidelines: diagno-
sis of tuberculosis in adults and children.
Clin Infect Dis 2017;​64(2):​e1-e33.
11. Banaei N, Gaur RL, Pai M. Interferon
gamma release assays for latent tubercu-
losis: what are the sources of variability?
J Clin Microbiol 2016;​54:​845-50.
12. Farhat M, Greenaway C, Pai M, Men-
zies D. False-positive tuberculin skin tests:
what is the absolute effect of BCG and
non-tuberculous mycobacteria? Int J Tuberc
Lung Dis 2006;​10:​1192-204.
13. Pai M, Denkinger CM, Kik SV, et al.
Gamma interferon release assays for de-
tection of Mycobacterium tuberculosis in-
fection. Clin Microbiol Rev 2014;​27:​3-20.
14. Pai M, Zwerling A, Menzies D. Sys-
tematic review: T-cell-based assays for the
diagnosis of latent tuberculosis infection:
an update. Ann Intern Med 2008;​149:​177-84.
15. Dorman SE, Belknap R, Graviss EA,

n engl j med 385;24  nejm.org  December 9, 2021 2279

The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.

et al. Interferon-γ release assays and tuber-
culin skin testing for diagnosis of latent
tuberculosis infection in healthcare work-
ers in the United States. Am J Respir Crit
Care Med 2014;​189:​77-87.
16. Sotgiu G, Saderi L, Petruccioli E, et al.
QuantiFERON TB Gold Plus for the diag-
nosis of tuberculosis: a systematic review
and meta-analysis. J Infect 2019;​79:​444-53.
17. Santin M, Muñoz L, Rigau D.
Interferon-γ release assays for the diag-
nosis of tuberculosis and tuberculosis in-
fection in HIV-infected adults: a system-
atic review and meta-analysis. PLoS One
2012;​7(3):​e32482.
18. Cattamanchi A, Smith R, Steingart KR,
et al. Interferon-gamma release assays for
the diagnosis of latent tuberculosis infec-
tion in HIV-infected individuals: a system-
atic review and meta-analysis. J Acquir
Immune Defic Syndr 2011;​56:​230-8.
19. Watkins RE, Brennan R, Plant AJ. Tu-
berculin reactivity and the risk of tuber-
culosis: a review. Int J Tuberc Lung Dis
2000;​4:​895-903.
20. Ledesma JR, Ma J, Zheng P, Ross JM,
Vos T, Kyu HH. Interferon-gamma release
assay levels and risk of progression to ac-
tive tuberculosis: a systematic review and
dose-response meta-regression analysis.
BMC Infect Dis 2021;​21:​467.
21. Comstock GW, Baum C, Snider DE Jr.
Isoniazid prophylaxis among Alaskan
Eskimos: a final report of the bethel iso-
niazid studies. Am Rev Respir Dis 1979;​
119:​827-30.
22. Hsu KH. Thirty years after isoniazid.
Its impact on tuberculosis in children and
adolescents. JAMA 1984;​251:​1283-5.
23. Akolo C, Adetifa I, Shepperd S, Vol-
mink J. Treatment of latent tuberculo-
sis infection in HIV infected persons.
Cochrane Database Syst Rev 2010;​ (1):​
CD000171.
24. Pape JW, Jean SS, Ho JL, Hafner A,
Johnson WD Jr. Effect of isoniazid pro-
phylaxis on incidence of active tuberculo-
sis and progression of HIV infection. Lan-
cet 1993;​342:​268-72.
25. Badje A, Moh R, Gabillard D, 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-e1089.
26. Sterling TR, Villarino ME, Borisov AS,
et al. Three months of rifapentine and
isoniazid for latent tuberculosis infection.
N Engl J Med 2011;​365:​2155-66.
2280 n engl j med 385;24  nejm.org  December 9, 2021
The New England Journal of Medicine
Downloaded from nejm.org at SUNY UPSTATE MEDICAL CENTER on January 27, 2022. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
Clinical Pr actice
27. Villarino ME, Scott NA, Weis SE, et al.
Treatment for preventing tuberculosis in
children and adolescents: a randomized
clinical trial of a 3-month, 12-dose regi-
men of a combination of rifapentine and
isoniazid. JAMA Pediatr 2015;​169:​247-55.
28. Sterling TR, Scott NA, Miro JM, et al.
Three months of weekly rifapentine and
isoniazid for treatment of Mycobacterium
tuberculosis infection in HIV-coinfected
persons. AIDS 2016;​30:​1607-15.
29. Martinson NA, Barnes GL, Moulton
LH, et al. New regimens to prevent tu-
berculosis in adults with HIV infection.
N Engl J Med 2011;​365:​11-20.
30. Belknap R, Holland D, Feng P-J, 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:​689-97.
31. Menzies D, Adjobimey M, Ruslami R,
et al. Four months of rifampin or nine
months of isoniazid for latent tuberculosis
in adults. N Engl J Med 2018;​379:​440-53.
32. Page KR, Sifakis F, Montes de Oca R,
et al. Improved adherence and less toxicity
with rifampin vs isoniazid for treatment
of latent tuberculosis: a retrospective
study. Arch Intern Med 2006;​166:​1863-70.
33. Lardizabal A, Passannante M, Kojakali
F, Hayden C, Reichman LB. Enhancement
of treatment completion for latent tuber-
culosis infection with 4 months of ri-
fampin. Chest 2006;​130:​1712-7.
34. Ronald LA, FitzGerald JM, Bartlett-
Esquilant G, et al. Treatment with isonia-
zid or rifampin for latent tuberculosis
infection: population-based study of hep-
atotoxicity, completion and costs. Eur
Respir J 2020;​55:​1902048.
35. Ena J, Valls V. Short-course therapy
with rifampin plus isoniazid, compared
with standard therapy with isoniazid, for
latent tuberculosis infection: a meta-analy-
sis. Clin Infect Dis 2005;​40:​670-6.
36. Swindells S, Ramchandani R, Gupta
A, et al. One month of rifapentine plus
isoniazid to prevent HIV-related tubercu-
losis. N Engl J Med 2019;​380:​1001-11.
37. Sterling TR, Njie G, Zenner D, et al.
Guidelines for the treatment of latent tu-
berculosis infection: recommendations
from the National Tuberculosis Control-
lers Association and CDC, 2020. MMWR
Recomm Rep 2020;​69:​1-11.
38. Treatment regimens for latent TB in-
fection (LTBI). Atlanta:​Centers for Dis-
ease Control and Prevention, 2020 (https://
www​.­cdc​.­gov/​­tb/​­topic/​­treatment/​­ltbi​.­htm).
39. International Union Against Tubercu-
losis Committee on Prophylaxis. Efficacy
of various durations of isoniazid preven-
tive therapy for tuberculosis: five years of
follow-up in the IUAT trial. Bull World
Health Organ 1982;​60:​555-64.
40. Comstock GW. How much isoniazid
is needed for prevention of tuberculosis
among immunocompetent adults? Int J
Tuberc Lung Dis 1999;​3:​847-50.
41. Sterling TR, Moro RN, Borisov AS, et al.
Flu-like and other systemic drug reactions
among persons receiving weekly rifapen-
tine plus isoniazid or daily isoniazid for
treatment of latent tuberculosis infection
in the PREVENT tuberculosis study. Clin
Infect Dis 2015;​61:​527-35.
42. Horsburgh CR Jr, Goldberg S, Bethel
J, et al. Latent TB infection treatment ac-
ceptance and completion in the United
States and Canada. Chest 2010;​137:​401-9.
43. Drain PK, Bajema KL, Dowdy D, et al.
Incipient and subclinical tuberculosis:
a clinical review of early stages and pro-
gression of infection. Clin Microbiol Rev
2018;​31(4):​e00021-e18.
44. Barry CE III, Boshoff HI, Dartois V,
et al. The spectrum of latent tuberculosis:
rethinking the biology and intervention
strategies. Nat Rev Microbiol 2009;​7:​845-
55.
45. Scriba TJ, Fiore-Gartland A, Penn-
Nicholson A, et al. Biomarker-guided
tuberculosis preventive therapy (CORTIS):
a randomised controlled trial. Lancet In-
fect Dis 2021;​21:​354-65.
46. WHO consolidated guidelines on tuber-
culosis. Module 1:​prevention — tubercu-
losis preventive treatment. Geneva:​World
Health Organization, 2020 (https://apps​
.­who​.­int/​­iris/​­bitstream/​­handle/​­10665/​
­331170/​­9789240001503​-­eng​.­pdf).
47. Tait DR, Hatherill M, Van Der Meeren
O, et al. Final analysis of a trial of M72/
AS01 E vaccine to prevent tuberculosis.
N Engl J Med 2019;​381:​2429-39.
48. Nemes E, Geldenhuys H, Rozot V, et al.
Prevention of M. tuberculosis infection with
H4:IC31 vaccine or BCG revaccination.
N Engl J Med 2018;​379:​138-49.
49. Gupta A, Hughes MD, Garcia-Prats
AJ, McIntire K, Hesseling AC. Inclusion of
key populations in clinical trials of new
antituberculosis treatments: current bar-
riers and recommendations for pregnant
and lactating women, children, and HIV-
infected persons. PLoS Med 2019;​16(8):​
e1002882.
Copyright © 2021 Massachusetts Medical Society.