Annals of Internal Medicine䊛

In the Clinic®

Tuberculosis Screening and Prevention

A
lthough tuberculosis (TB) rates remain
low in the United States, 10.4 million
people were diagnosed and 1.4 million Diagnosis
died of the disease worldwide in 2015. The past
decade has seen major advances in prevention,
diagnosis, and treatment. New diagnostics to
Treatment
detect latent TB infection (LTBI) and clinical dis-
ease have been developed, and new drugs,
particularly for drug-resistant TB, have become
increasingly available. Despite these advances, Tool Kit
in 2014 TB surpassed HIV/AIDS as the deadliest
infectious disease in the world (1), and it contin-
ues to be the leading killer of persons infected Patient Information
with HIV globally.

The CME quiz is available at Annals.org. Complete the quiz to earn up to 1.5 CME credits.

Physician Writer doi:10.7326/AITC201702070

Karen R. Jacobson, MD,
MPH CME Objective: To review current evidence for screening and prevention, diagnosis, and
treatment of tuberculosis.
Funding Source: American College of Physicians.
Acknowledgment: The author thanks Elizabeth J. Ragan for administrative and technical
support on this manuscript and Dr. Patricio Escalante, writer of the earlier version.
With the assistance of additional physician writers, the editors of Annals of Internal Medicine
develop In the Clinic using MKSAP and other resources of the American College of
Physicians.
In the Clinic does not necessarily represent official ACP clinical policy. For ACP clinical
guidelines, please go to https://www.acponline.org/clinical_information/guidelines/.
Disclosures: Dr. Jacobson, ACP Contributing Author, has disclosed no conflicts of interest.
The form can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms
.do?msNum=M16-2138.

© 2017 American College of Physicians

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017


Screening and Prevention
1. World Health Organiza-
tion. Global Tuberculosis
Report 2016. Geneva:
World Health Organiza-
tion; 2016.
2. Centers for Disease Con-
trol and Prevention. Fact
Sheet: Trends in tubercu-
losis. 2014. Accessed at
www.cdc.gov/tb
/publications/factsheets
/statistics/tbtrends.htm
on 20 December 2016.
3. World Health Organiza-
tion. The END TB Strategy:
Global strategy and tar-
gets for tuberculosis pre-
vention, care and control
after 2015. Geneva: World
Health Organization;
2014. Accessed at www
.who.int/tb/strategy
/End_TB_Strategy.pdf
?ua=1 on 20 December
2016.
4. Alvarez GG, Van Dyk DD,
Davies N, Aaron SD, Cam-
eron DW, Desjardins M,
et al. The feasibility of the
interferon gamma release
assay and predictors of
discordance with the tu-
berculin skin test for the
diagnosis of latent tuber-
culosis infection in a re-
mote aboriginal commu-
nity. PLoS One. 2014;9:
e111986. [PMID:
25386908]
5. Ribeiro-Rodrigues R, Kim
S, Coelho da Silva FD,
Uzelac A, Collins L, Palaci
M, et al. Discordance of
tuberculin skin test and
interferon gamma release
assay in recently exposed
household contacts of
pulmonary TB cases in
Brazil. PLoS One. 2014;9:
e96564. [PMID:
24819060] http://dx.doi
.org/10.1371/journal.pone
.0096564 on 20 Decem-
ber 2016.
6. European Centre for Dis-
ease Prevention and Con-
trol. Management of con-
tacts of MDR TB and XDR
TB patients. Stockholm:
European Centre for Dis-
ease Prevention and Con-
trol; 2012.
姝 2017 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
In the United States, 9421 TB
cases (3 cases per 100 000 per-
sons) were reported in 2014, a
69% decrease since a resurgence
in 1992 (2). Sixty-six percent of
these cases occurred in foreign-
born persons, a rate of infection
13 times that of U.S.-born per-
sons (2). Racial and ethnic
groups are disproportionately
affected—the highest rates of
Who should be screened for
latent TB infection?
TB is caused by the acid-fast ba-
cillus (AFB) Mycobacterium tu-
berculosis, which is transmitted
through airborne respiratory
droplets from the throat and
lungs of persons with active re-
spiratory disease. When M tuber-
culosis droplets are inhaled and
delivered to the terminal airways,
macrophages ingest the myco-
bacteria, which continue to multi-
ply intracellularly and can poten-
tially spread to other organs
through the lymphatic system
and bloodstream. Most TB-
infected persons remain asymp-
tomatic and enter the latency
phase (LTBI). They have a 10%
lifetime risk for progressing from
latent to active TB; half of that risk
is progression during the first 2
years after infection (“primary
progression”).
The goal of LTBI testing is to
identify and treat persons at in-
creased risk for TB. These per-
sons include contacts of an indi-
vidual with known infectious
pulmonary TB, those at high risk
for potential exposure, and those
at high risk for rapid disease pro-
gression regardless of known TB
contact (see the Box). Risk for
primary progression in the first 2
years after infection is age-
dependent— children younger
than 2 years have the highest
risk. Thus, particular attention
should be paid to LTBI screening
ITC18 In the Clinic
infection occur among Asians as
well as native Hawaiians and
other Pacific
Islanders. Building on the mo-
mentum of new scientific dis-
coveries, the World Health Or-
ganization (WHO) published the
END TB Strategy in 2014, which
aims to reduce TB incidence
and deaths by 90% and 95%,
respectively, by 2035 (3).
and treatment in young children
(4, 5).
If the results of the first test after
exposure of close contacts of a
known case patient with TB (per-
sons from the same household
sharing common habitation
rooms, or those with prolonged
and frequent exposure in such
settings as workplaces, schools,
prisons, hospitals, and some so-
cial settings) (6) are negative, the
test should be repeated in 8 –12
weeks to confirm lack of conver-
sion to positive results.
What tests are used to screen
for TB infection?
The 2 tests used to detect TB are
the Mantoux tuberculin skin test
(TST) and interferon-␥–release
assays (IGRAs). Neither test can
distinguish between latent and
active disease, and neither is ap-
propriate as the sole investiga-
tion for active disease (Table 1).
Both tests are based on cell-
mediated immunity and thus are
more likely to yield false-negative
results in persons with advanced
immunosuppression.
The TST has been used for more
than 100 years. Purified protein
derivative (PPD) is injected intra-
dermally (usually into the inner
surface of the forearm). The pa-
tient returns in 48 –72 hours, and
the test is read by assessing the
transverse diameter of induration
(not erythema). Positive results
indicate a delayed-type hyper-
Annals of Internal Medicine 7 February 2017
 sensitivity response mediated by
T lymphocytes. To increase the
specificity of the test, criteria for
positivity are based on the pa-
tient's risk factors for M tubercu-
losis infection (Table 1).
The largest threshold (≥15 mm
induration) has the lowest sensi-
tivity but the highest specificity
and is therefore useful to evalu-
ate patients in regions with high
prevalence of nontuberculous
mycobacterium rather than
M tuberculosis, such as the
United States. False-negative re-
sults are more common in pa-
tients with recent TB infection,
age younger than 6 months,
overwhelming active TB, recent
vaccination with a live virus (e.g.,
measles), recent viral infection
(e.g., measles or varicella), and
anergy. False-positive results can
be caused by a history of bacille
Calmette-Guérin (BCG) vaccina-
tion or infection with nontubercu-
lous mycobacterium. People who
received the BCG vaccine should
have their TST interpreted ac-
cording to standard criteria, al-
though recent BCG vaccination
can be an indication for IGRA
testing instead. For patients who
received the BCG vaccine when
they were younger than 1 year,
no discernible effect on TST re-
mains after 10 or more years (7).
Patients with remote exposure to
TB may initially have a negative
TST result that can become posi-
tive several weeks later. This is
called the “booster effect” and is
considered a true-positive result.
It is more common in elderly per-
sons, although it may also occur
more often in persons with a his-
tory of BCG or nontuberculous
mycobacteria. For this reason,
current guidelines recommend a
2-step initial TST for all health
care workers, with the second
test done on the opposite fore-
arm 7–21 days after initial nega-
tive results (8). Two-step testing
7 February 2017 Annals of Internal Medicine
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
is not required when IGRA is
used (9).
The Centers for Disease Control
and Prevention (CDC) now en-
dorses IGRAs, including the
QuantiFERON-TB Gold In-Tube
(QFT) (Qiagen) and the
T-SPOT.TB (Oxford Immunotec),
in nearly all clinical settings
where TST is recommended (8,
10). One exception is children
younger than 5 years, for whom
some experts recommend both
tests for increased specificity.
IGRAs indicate sensitization to
M tuberculosis by measuring the
release of interferon-␥ in the
blood by T cells as a response to
M tuberculosis–associated anti-
gen and do not require a second
patient visit. The initially higher
cost of IGRAs should be taken
into account. These tests are con-
sidered to be as sensitive as but
more specific than TSTs because
persons who have other nontu-
berculous mycobacterium infec-
tions or have had BCG do not
have positive results. IGRAs are
preferred to TSTs in persons who
received BCG as treatment for
cancer or as a vaccine and in those
who are less likely to return for a
test reading in 48 –72 hours (11).
Once a person has a positive TST
or IGRA result, repeated testing
has no clinical utility. Unless sus-
picion for error in interpretation
or performance of the first test is
high, a positive result should be
considered a true positive. In a
patient with a known positive TST
result, repeated testing can result
in a more severe, and at times ne-
crotic, inflammatory response (12).
Two challenges for providers are
how to interpret discordance be-
tween results on TST and IGRA
and how to interpret reversions if
a test is repeated. Historically,
stable TST converters (i.e., those
who do not revert with later test-
ing) had significantly larger areas
of induration than reverters at the
In the Clinic
Risk Factors for TB Infection
or Progression to Disease
After Infection
Persons with HIV infection
Patients with other
immunocompromising
conditions, including organ
transplant recipients and
others with illnesses that
require immunosuppressive
therapy (the equivalent of
≥15 mg of prednisone per
day for ≥4 wk, which
includes most persons
receiving tumor necrosis
factor-␣ antagonists)
Persons recently in close contact
with a person with active TB
Persons with fibrotic changes on
chest radiography consistent
with old TB
Recent (<5 y) arrivals from
high-prevalence countries
Mycobacteriology laboratory
personnel
Residents or employees in
high-risk settings, such as
prisons and jails, nursing
homes and other long-term
facilities for elderly persons,
hospitals and other health
care facilities, residential
facilities for patients with
AIDS, and homeless shelters
Injection drug users, users of
other drugs (e.g., crack
cocaine), and tobacco
smokers
Persons with clinical conditions
that put them at high risk for
active disease (including
diabetes mellitus; silicosis,
intestinal bypass, gastrectomy,
or other type of bariatric
surgery; cancer of the head
and neck; chronic
malabsorption syndromes;
end-stage renal disease;
hematologic cancer; and
body weight ≥10% below
ideal level)
Children aged ≤4 years exposed
to adults with TB
ITC19 姝 2017 American College of Physicians
 Table 1. Interpretation of Tuberculin Skin Test Results: Criteria for Tuberculin Positivity by Risk Group,
According to Size of Induration, in Millimeters*
≥5 ≥10 ≥15
HIV-positive persons Recent (<5 y) arrivals from high-prevalence TB countries Persons with no risk
Recent contacts of persons with active Injection drug users factors for TB
TB Residents or employees of high-risk congregate settings,
Persons with fibrotic changes on such as prisons and jails, nursing homes and other
chest X-ray consistent with old TB long-term facilities for elderly persons, health care
Patients who have had organ facilities, residential facilities for AIDS patients,
transplantation and other homeless shelters
immunosuppressive conditions Persons with comorbid conditions that place them at
(receiving equivalent of ≥15 mg high risk for TB (silicosis, diabetes mellitus, severe
prednisone/d for >4 wk) kidney disease, certain types of cancer, some intestinal
conditions)
Mycobacteriology laboratory personnel
Children aged ≤4 y

TB = tuberculosis.
* Data from reference 2.

time of TST conversion (13), mak- What can patients do to

ing the larger threshold in most decrease the likelihood of
U.S.-based populations more infecting others?
likely to be true-positive. More TB patients have been shown to
concerning has been that rates of infect up to 10 –15 other people
serial conversion from negative over the course of a year. How-
to positive IGRA results among ever, once a patient has received
health care workers in North effective therapy for 2 weeks, risk
America have been reported at for transmission is nearly, if not
nearly an order of magnitude completely, eliminated (19 –21).
higher than historical or concur- Therefore, prompt identification
rent TST conversion rates (14 – of patients with active TB is criti-
16), accompanied by high rates cal to prevent transmission.
(>60%) of reversion on subse-
quent testing. Prevention of spread in hospitals
includes respiratory isolation of
7. Farhat M, Greenaway C,
Pai M, Menzies D. False- Hypotheses for IGRA variability in patients with confirmed or sus-
positive tuberculin skin a single patient have included pected TB (22). Staff who come in
tests: what is the absolute
effect of BCG and non- that initial conversion followed by contact with an infectious or sus-
tuberculous mycobacteria?
Int J Tuberc Lung Dis.
reversion may reflect exposure pected infectious patient should
2006;10:1192-204. that does not lead to persistent wear previously fitted particulate
[PMID: 17131776]
8. Centers for Disease Con- infection (17); operational rea- respirators (N95 masks). Patients
trol and Prevention. Latent sons for variability in results on
Tuberculosis Infection: A suspected of having TB can be
Guide for Primary Health the QFT, including manufactur- removed from isolation after 3
Care Providers. 2014.
Accessed at www.cdc.gov ing and processing issues; ana- sputum smears, collected at least
/tb/publications lytic variability; and host immuno- 8 hours apart, are confirmed to
/ltbi/diagnosis.htm on
20 December 2016. logic changes over time (18). be negative and an alternative
9. QIAGEN. Frequently Asked
Questions: QuantiFERON® Recent studies have also found diagnosis has been made (22).
-TB Gold. 2013. Accessed evidence of positive QFT results Patients with confirmed TB can
at http://usa.quantiferon
.com/irm/content/pdfs after TST in the prior 6 months, usually be removed from isolation
/FAQ_QFT_HCP-US_EN
_1113_H_LR.pdf on 20
so such proximate dual testing once they begin adequate ther-
December 2016. should be particularly avoided apy; show a significant clinical re-
10. Mazurek GH, Jereb J,
Vernon A, LoBue P, Gold- (16, 17). sponse; and are negative for AFB
berg S, Castro K. Up-
dated guidelines for
on 3 sputum smears, also ob-
using interferon gamma
For these reasons, providers tained at least 8 hours apart (22).
release assays to detect should use the LTBI test that
Mycobacterium tubercu-
losis infection—United best fits their patient's needs Patients who are not too ill can
States, 2010. MMWR and generally should use only be sent home after initiation of
Recomm Rep. 2010;59:
1-25. [PMID: 20577159] one. therapy, although they must be

姝 2017 American College of Physicians ITC20 In the Clinic Annals of Internal Medicine 7 February 2017

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 isolated from outsiders. Healthy
adults residing in the home have
probably been exposed for
weeks to months before diagno-
sis of the infected patient and are
unlikely to benefit from isolation
(23). However, if there are high-
risk contacts in the household
(i.e., children aged <5 years or
immunocompromised persons),
TB patients should reside else-
where until they meet criteria for
noninfection (22).
The BCG vaccine is not used in the
United States. However, knowl-
edge of whether persons born in
other countries received the vac-
cine, at what age, and whether
they received a booster is particu-
larly helpful when deciding which
LTBI diagnostic test to use. An in-
teractive Web site lists BCG vac-
cine practices in 180 countries
(www.bcgatlas.org) (24).
For patients with latent TB
infection, what can be done to
decrease the likelihood of
developing active disease?
Patients with evidence of LTBI
(i.e., positive results on IGRA or
TST), in whom active TB has been
excluded, should be offered pro-
phylaxis. Patients at greater risk
for active disease should be par-
ticularly encouraged to take pro-
phylaxis, including those with HIV
infection, close contacts of per-
sons with active TB in the past 2
years (especially infants and chil-
dren aged <5 years), those who
have fibrotic changes on a chest
radiograph consistent with old
TB, those with medical conditions
known to increase risk (e.g., dia-
betes mellitus; silicosis; organ
transplantation; immunosuppres-
sive therapy, such as the equiva-
lent of >15 mg of prednisone per
day for ≥1 month; and receipt of
tumor necrosis factor-␣ antago-
nists), those who inject illicit
drugs and other high-risk sub-
stance users, and residents and
employees in high-risk congregate
settings (e.g., correctional facilities,
7 February 2017 Annals of Internal Medicine
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
nursing homes, homeless shelters,
and hospitals) (25, 26).
Immunocompetent children
younger than 5 years should be
started on “window prophylaxis,”
regardless of baseline TST/IGRA
status, and this therapy should be
stopped only if TST/IGRA results are
negative 8–10 weeks after the last
contact with a contagious case (27).
The CDC endorses 4 LTBI treat-
ment regimens: 9 or 6 months of
isoniazid (INH), 3 months of INH
plus rifapentine (RPT), or 4
months of rifampin (RIF) (Table
2). The newest, the “12-dose reg-
imen,” is a combination of INH
and RPT given weekly for 3
months; treatment completion
was 82% among patients receiv-
ing the 12-dose regimen com-
pared with 69% among those
receiving 9 months of INH in the
TB Trials Consortium Prevent TB
study (P < 0.001) (28). Directly
observed therapy (DOT) for this
regimen is currently recom-
mended because the impact of
missed doses is unknown, but at
least 1 CDC-sponsored study
suggests that self-administration
does not lead to inferior adher-
ence in the United States (29).
DOT also provides the opportu-
nity to screen patients for ad-
verse effects and to stop therapy
if they occur. A meta-analysis re-
ports that the weekly INH–RPT
regimen had higher treatment
completion and less hepatotoxic-
ity but more frequent treatment-
limiting adverse events com-
pared with 9 months of daily
treatment with INH (30). The 12-
dose regimen is an option for
otherwise-healthy persons aged
12 years or older. It is not recom-
mended for children younger
than 12 years, HIV-infected per-
sons receiving antiretroviral med-
icines (due to drug interactions),
pregnant women or women who
expect to become pregnant dur-
ing treatment, persons infected
with drug-resistant strains (31),
In the Clinic
11. Ringshausen FC,
Schablon A, Nienhaus A.
Interferon-gamma re-
lease assays for the tu-
berculosis serial testing
of health care workers: a
systematic review. J
Occup Med Toxicol.
2012;7:6. [PMID:
22537915]
12. Bunnet D, Kerleguer A,
Kim P, Pean P, Phuong
V, Heng N, et al. Necrotic
tuberculin skin (Man-
toux) test reaction: a case
report and an estimation
of frequency. Chest.
2015;148:e1-4. [PMID:
26149555]
13. Johnson DF, Malone LL,
Zalwango S, Mukisa
Oketcho J, Chervenak
KA, Thiel B, et al; Tuber-
culosis Research Unit.
Tuberculin skin test re-
version following isonia-
zid preventive therapy
reflects diversity of im-
mune response to pri-
mary Mycobacterium
tuberculosis infection.
PLoS One. 2014;9:
e96613. [PMID:
24796677]
14. Zwerling A, Benedetti A,
Cojocariu M, McIntosh F,
Pietrangelo F, Behr MA,
et al. Repeat IGRA testing
in Canadian health work-
ers: conversions or unex-
plained variability? PLoS
One. 2013;8:e54748.
[PMID: 23382955]
15. Slater ML, Welland G, Pai
M, Parsonnet J, Banaei
N. Challenges with
QuantiFERON-TB Gold
assay for large-scale,
routine screening of U.S.
healthcare workers. Am J
Respir Crit Care Med.
2013;188:1005-10.
[PMID: 23978270]
16. Dorman SE, Belknap R,
Graviss EA, Reves R,
Schluger N, Weinfurter
P, et al; Tuberculosis
Epidemiologic Studies
Consortium. Interferon-␥
release assays and tuber-
culin skin testing for
diagnosis of latent tuber-
culosis infection in
healthcare workers in the
United States. Am J
Respir Crit Care Med.
2014;189:77-87. [PMID:
24299555]
17. Andrews JR, Hatherill M,
Mahomed H, Hanekom
WA, Campo M, Hawn TR,
et al. The dynamics of
QuantiFERON-TB gold
in-tube conversion and
reversion in a cohort of
South African adoles-
cents. Am J Respir Crit
Care Med. 2015;191:
584-91. [PMID:
25562578]
18. Pai M, Denkinger CM,
Kik SV, Rangaka MX,
Zwerling A, Oxlade O,
et al. Gamma interferon
release assays for detec-
tion of Mycobacterium
tuberculosis infection.
Clin Microbiol Rev.
2014;27:3-20. [PMID:
24396134]
ITC21 姝 2017 American College of Physicians
 Table 2. Treatment Regimens for Latent TB*
Drugs Duration Interval Comments Rating
(evidence)†

HIV– HIV+
INH 9 mo Daily Preferred treatment for: A2 A2
• Persons with HIV
• Children aged 2–11 y
• Pregnant women (with pyridoxine/vitamin B6 supplements)‡
Twice weekly§ Preferred treatment for pregnant women (with pyridoxine/vitamin B2 B2
B6 supplements)‡
INH 6 mo Daily Not indicated for HIV-infected persons, persons with fibrotic lesions B1 C1
on chest radiographs, or children
Twice weekly§ B2 C1
INH + RPT 3 mo Once weekly§ Treatment for persons ≥12 y A1 A兩兩1
Not recommended for persons:
• <2 y
• Receiving antiretroviral treatment
• Presumed to be infected with INH- or RIF-resistant
Mycoplasma tuberculosis
• Who are pregnant or expect to become pregnant
RIF 4 mo Daily Not recommended for persons: B2 B3
• Receiving medications that interact with rifamycins
• Who wear contact lenses
• Who are pregnant or expect to become pregnant

INH = isoniazid; RIF = rifampin; RPT = rifapentine; TB = tuberculosis.

* Adapted from Centers for Disease Control and Prevention. Treatment Regimens for Latent TB Infection (LTBI) [Internet]. 2016.
Available from www.cdc.gov/tb/topic/treatment/ltbi.htm.
† A = preferred; B = acceptable alternative; C = offer if A or B are not available; 1 = randomized, clinical trial; 2 = data from
clinical trials that were not randomized or were conducted in other populations; 3 = expert opinion.
‡ May be delayed until after delivery and cessation of breastfeeding except in cases of increased risk for placental infection or
progression to active TB.
§ Using directly observed therapy.
兩兩 In otherwise-healthy HIV patients not receiving antiretroviral medication.

19. Rouillon A, Perdrizet S,

Parrot R. Transmission of and those receiving medications alternative multidrug prophylaxis
tubercle bacilli: the ef- that interact with rifamycins. regimen if a person is at high risk
fects of chemotherapy.
Tubercle. 1976;57:275- for progression to active disease
99. [PMID: 827837] In pregnant women, a 6- to (i.e., HIV-positive) and the suscepti-
Accessed at http://dx.doi
.org/10.1016/S0041 9-month course of INH may be bility of the infecting organism is
-3879(76)80006-2 on 20 delayed until after delivery and known (33). Such patients should
December 2016.
20. Riley RL, Mills CC, cessation of breastfeeding, ex- be referred to an experienced
O’Grady F, Sultan LU,
Wittstadt F, Shivpuri DN. cept when risk for placental infec- specialist.
Infectiousness of air from tion or progression to active TB is
a tuberculosis ward.
Ultraviolet irradiation of increased (i.e., immunocompro- How can patients with active
infected air: comparative TB protect household members
infectiousness of differ- mising conditions, such as HIV
ent patients. Am Rev co-infection or recent M tubercu- and other contacts from infection?
Respir Dis. 1962;85:511-
25. [PMID: 14492300] losis infection) (32). For those Patients who are infected with or are
21. Dharmadhikari AS,
Mphahlele M, Venter K, exceptions, treatment with INH suspected of being infected with TB
Stoltz A, Mathebula R, plus pyridoxine/vitamin B6 (plus must be educated to cover their
Masotla T, et al. Rapid
impact of effective treat- B6 to the newborn) should be mouth and nose when they cough
ment on transmission of
multidrug-resistant tu- initiated during pregnancy or or sneeze, not sleep in a room with
berculosis. Int J Tuberc breastfeeding with close moni- other household members, and
Lung Dis. 2014;18:
1019-25. [PMID: toring for INH hepatotoxicity, refrain from having home visitors
25189547]
22. Jensen PA, Lambert LA, which is more common in until they are noninfectious.
Iademarco MF, Ridzon R. women during pregnancy.
Guidelines for prevent- What are the physician's public
ing the transmission of
Mycobacterium tubercu- Treatment of LTBI in contacts of health responsibilities after
losis in health-care set-
tings, 2005. MMWR patients with drug-resistant TB is diagnosing active TB?
Recomm Rep. 2005;54:
1-141. [PMID:
not well-studied. Current guide- All 50 states require health care
16382216] lines recommend considering an providers to notify the TB pro-

姝 2017 American College of Physicians ITC22 In the Clinic Annals of Internal Medicine 7 February 2017

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 gram in their respective depart- to set up processes to screen po-
ments of health within 24 hours tential contacts, and take steps to
in cases of suspected or con- avoid further transmission. Pro-
firmed active TB (34); require- viders must report treatment
ments for further reporting vary progress as well as patients who 23. Kamat SR, Dawson JJ,
by state. Within 24 hours, a are nonadherent, leave medical Devadatta S, Fox W,
Janardhanam B, Rad-
state's TB program is responsible facilities against medical advice, hakrishna S, et al. A
for notifying the local board of or continue to place others at controlled study of the
influence of segregation
health of the patient's residence, risk. of tuberculous patients
for one year on the attack
rate of tuberculosis in a
5-year period in close
family contacts in South
Screening and Prevention... Clinicians should screen close contacts of India. Bull World Health
a person with active pulmonary TB or those who are at high risk for in- Organ. 1966;34:517-32.
fection or progression to disease once infected. Clinicians should pre- [PMID: 5296379]
24. Zwerling A, Behr MA,
vent infection by identifying and treating persons with active pulmonary Verma A, Brewer TF,
TB. Patients with evidence of infection but not disease should be of- Menzies D, Pai M. The
BCG World Atlas: a data-
fered LTBI therapy. Patients suspected of having TB should be placed base of global BCG vacci-
on airborne isolation, and staff who provide care should wear particu- nation policies and prac-
late respirators. Clinicians should notify public health authorities about tices. PLoS Med. 2011;8:
e1001012. [PMID:
patients with suspected active TB. 21445325]
25. Centers for Disease Con-
trol and Prevention.
Deciding When to Treat
CLINICAL BOTTOM LINE Latent TB Infection.
2016. Accessed at www
.cdc.gov/tb/topic
/treatment/decideltbi
.htm on 20 December
2016.
Diagnosis 26. Smieja MJ, Marchetti CA,
Cook DJ, Smaill FM.
Isoniazid for preventing
What are the signs and picion for active TB should be tuberculosis in non-HIV
symptoms of active TB? high in patients with HIV who infected persons.
Cochrane Database Syst
Table 3 shows some of the main have been exposed to TB. Active Rev. 2000:CD001363.

findings from the history and
physical examination that are as-
sociated with active TB. Pulmo-
nary disease is overwhelmingly
the most common presentation
(>80%) (35). Persons with com-
promised immune systems (e.g.,
HIV co-infected patients) may
present with few classic symp-
toms and are more likely to pres-
ent with extrapulmonary TB.
What are the signs and
symptoms of active TB among
HIV co-infected patients?
Signs and symptoms vary by de-
gree of immunodeficiency. Pa-
tients with CD4 counts greater
than 350 per cubic millimeter are
more likely to present with classic
constitutional symptoms; those
with CD4 counts less than 200
per cubic millimeter are more
likely to have atypical and asymp-
tomatic presentation and ex-
trapulmonary disease (36, 37).
For this reason, the index of sus-
microbiological screening for TB, [PMID: 10796642]
27. Loeffler AM. Pediatric
regardless of whether symptoms Tuberculosis: An Online
Presentation. 2010.
are present, is important because Accessed at www
disease can be missed when .currytbcenter.ucsf.edu
/products/pediatric
screening is based on signs and -tuberculosis-online
symptoms alone. Up to 25% of -presentation?productID
=ONL-10 on 20 Decem-
patients screened before starting ber 2016.
28. Sterling TR, Villarino ME,
antiretroviral drugs in South Af- Borisov AS, Shang N,
rica were found to have microbi- Gordin F, Bliven-
Sizemore E, et al; TB
ologically confirmed TB (38, 39). Trials Consortium PRE-
VENT TB Study Team.
How is active TB diagnosed? Three months of rifapen-
tine and isoniazid for
In resource-rich settings, patients latent tuberculosis infec-
tion. N Engl J Med.
with symptoms, signs, or radio- 2011;365:2155-66.
graphic findings suggestive of [PMID: 22150035] www
active TB should receive 3 serial
.ncbi.nlm.nih.gov
/pubmed/22150035 on
sputum microscopy and myco- 20 December 2016.
29. Belknap R, Borisov A,
bacterial cultures collected at Holland D, Feng P-J,
least 8 hours apart, in addition to Millet J-P, Martinson N,
et al. Adherence to Once-
chest radiography. Sputum in- Weekly Self-
Administered INH and
duction should be done if pa- Rifapentine for Latent TB:
tients cannot expectorate sponta- iAdhere. In: Conference
on Retroviruses and
neously. If there are signs of Opportunistic Infections.
Seattle; 2017. Accessed
extrapulmonary disease, samples at www.croiconference
from those areas, such as lymph .org/sessions/adherence
-once-weekly-self
nodes or the pleural space, -administered-inh-and
should also be evaluated. Histo- -rifapentine-latent-tb
-iadhere on 20
pathologic evidence of caseating December 2016.

7 February 2017 Annals of Internal Medicine In the Clinic ITC23 姝 2017 American College of Physicians

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 Table 3. Findings From the History and Physical Examination in Patients With Active TB
History Notes
TB exposure, infection, disease, or treatment Patients with recent exposure to TB are more likely to develop disease;
however, known exposure is not necessary to have TB. Clinicians may
contact their local health department to learn whether a patient has received
TB treatment in the past and what regimen was used
HIV infection or other medical conditions, or HIV-infected patients with latent TB infection have >20-fold risk for progression to
demographic factors that may increase the active TB than patients without HIV. Signs and symptoms of other concurrent
risk for, or change the presentation of, TB medical conditions may mask signs and symptoms of TB
Fever Less likely to be present in elderly persons. Absence of fever does not rule out
TB. Patients may report feeling feverish without having fever
Malaise
Night sweats A classic symptom of TB but may only be present in disease of long duration
Weight loss More common in advanced, extrapulmonary, or disseminated disease
Gynecologic symptoms Pelvic pain, menstrual irregularities, and infertility are the most common

Physical Examination
Systemic signs Fever, wasting, hepatomegaly, pulmonary findings, lymphadenopathy, and
splenomegaly can be present
Throat examination Hoarseness
Lymph node examination May be palpable with pulmonary, disseminated disease, or scrofula
Pulmonary examination Generally not helpful but may include rales, signs of consolidation, or findings
consistent with (often unilateral) pleural effusion (including pleuritic pain)
Pericardial disease Tachycardia, increased venous pressure, hepatomegaly, pulsus paradoxus,
and friction rub
Abdominal examination Ascites, “doughy” abdomen, or abdominal mass. Hepatosplenomegaly in
disseminated disease
Genitourinary examination Recurrent urinary tract infection with no organisms on culture. In men, beaded
vas deferens on palpation, draining scrotal sinus, epididymitis or induration
of prostrate or seminal vesicles
Musculoskeletal examination Joint swelling, gibbus deformity, or localized pain
Neurologic examination Abnormal behavior, headache, seizure

TB = tuberculosis.

granulomas is suggestive of ac- Bronchoscopy with either bron-

30. Sharma SK, Sharma A,
Kadhiravan T, Tharyan P.
Rifamycins (rifampicin,
rifabutin and rifapentine)
compared to isoniazid for
preventing tuberculosis
in HIV-negative people at
risk of active TB.
Cochrane Database Syst
Rev. 2013:CD007545.
[PMID: 23828580]
31. Castro KG. New Regimen
Makes Treating Latent
Tuberculosis Infection
Easier. Medscape. 2012.
32. Sackoff JE, Pfeiffer MR,
Driver CR, Streett LS,
Munsiff SS, DeHovitz JA.
Tuberculosis prevention
for non-US-born preg-
nant women. Am J Ob-
stet Gynecol. 2006;194:
451-6. [PMID:
16458645]
33. Management of persons
exposed to multidrug-
resistant tuberculosis.
MMWR Recomm Rep.
1992;41:61-71. [PMID:
1640921]
tive TB but is not diagnostic; simi-
larly, AFB positivity alone does
not confirm diagnosis.
The gold standard for TB diagno-
sis is a positive mycobacterial
culture from liquid medium. Neg-
ative results on AFB smears occur
in half of patients with active pul-
monary TB and in an even higher
percentage of HIV co-infected
persons; therefore, mycobacte-
rial cultures should be routinely
done, even in patients with nega-
tive smears. In persons sus-
pected of having disseminated
TB or who are highly immuno-
suppressed (i.e., advanced AIDS),
blood culture and collection of
first-void urine for mycobacterial
analysis can be done.
choalveolar lavage or biopsy
should be used for diagnosis in
patients who are suspected of hav-
ing active disease but have nega-
tive sputum samples (40). Symp-
toms similar to those of TB, such as
weight loss, fever, and cough, can
be caused by nontuberculous my-
cobacteria, aspiration pneumonia,
lung abscesses, Wegener granulo-
matosis, actinomycosis, and can-
cer. Thus, TB infection should be
confirmed on culture whenever
possible. Clinicians should be
aware of other diseases that may
resemble TB on pathologic exami-
nation, including nontuberculous
mycobacteria, syphilis, brucellosis,
sarcoidosis, or such fungal dis-
eases as histoplasmosis or coccid-
iomycosis (41).

姝 2017 American College of Physicians ITC24 In the Clinic Annals of Internal Medicine 7 February 2017

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 A major operational challenge in
diagnosing TB is the lag time un-
til results from solid media (i.e.,
Löwenstein–Jensen medium) are
available. Liquid medium (e.g.,
Mycobacteria Growth Indicator
Tube) has a shorter median time
to positivity for both smear-
positive (7 vs. 14 days) and
smear-negative (14 vs. 25 days)
disease (42). The CDC has en-
dorsed the additional use of nu-
cleic acid amplification (NAA)
testing on sputum specimens
when TB is suspected (43). This
test can provide results in 24 – 48
hours and has a greater positive
predictive value for true M tuber-
culosis (>95%) compared with
AFB smear-positive specimens in
settings where nontuberculous
mycobacterium are common.
NAA can also rapidly confirm the
presence of M tuberculosis in
50%– 80% of AFB smear-
negative, culture-positive speci-
mens. Compared with culture,
results are available weeks earlier
for 80%–90% of patients with sus-
pected pulmonary TB among
those in whom it is ultimately con-
firmed by culture (44). Thus, al-
though NAA should not replace
culture, it can facilitate earlier deci-
sion making regarding whether to
initiate treatment for TB.
The Xpert MTB/RIF assay is an
NAA test that uses a disposable
cartridge in the GeneXpert In-
strument System, and results are
available within 2 hours. This as-
say has been reported to have
specificity of 99% and sensitivity
of 98% in smear-positive and
specificity of 99% and sensitivity
of 67% in smear-negative sputum
specimens (45).
Xpert MTB/RIF testing has been
widely implemented in resource-
limited settings with high TB
prevalence (46). In a large U.S.
study, diagnostic performance of
this test was similar to that in
high–TB burden settings, sup-
porting its use for initial evalua-
7 February 2017 Annals of Internal Medicine
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
tion of suspected TB in conjunc-
tion with other tests (47). Two
negative Xpert MTB/RIF tests
may in the future replace the cur-
rent standard of 3 sputum smears
to rule out infectious TB, permit-
ting removal of respiratory pre-
cautions. A result indicating RIF
susceptibility can also reassure
providers that the patient is un-
likely to have multidrug resis-
tance (MDR).
How should drug susceptibility
testing be done, and in which
patients?
If active TB is confirmed, suscep-
tibility testing for first-line drugs
should be done, ideally on the
initial, pretreatment isolate.
Drug-resistant TB was previously
believed to be caused by ac-
quired resistance in initially drug-
susceptible disease due to poor-
quality drugs, poor adherence,
or a weak drug combination.
However, a globally increasing
proportion of drug-resistant
cases is now occurring due to
person-to-person transmission.
Thus, the ability to predict who
has drug-resistant disease has
weakened, making baseline re-
sistance testing even more
important.
Culture remains the standard for
drug susceptibility testing and
should be done at a reference
laboratory. If available, Xpert
MTB/RIF can detect RIF
resistance, allowing for more
rapid triage of samples for addi-
tional testing. If RIF resistance is
detected, drug susceptibility test-
ing for both first-line and second-
line drugs should be done (see
the CDC TB Web site for details)
and the treatment regimen
should be adjusted accordingly.
Drug-resistant TB refers to M tu-
berculosis with resistance to any
of the TB drugs. An important
distinction is MDR TB, which is
defined as resistance to INH and
RIF. Patients with this type of
In the Clinic
34. Centers for Disease Con-
trol and Prevention.
Menu of Suggested
Provisions For State
Tuberculosis Prevention
and Control Laws. 2010.
Accessed at www.cdc.gov
/tb/programs/laws/menu
/caseid.htm on
20 December 2016.
35. Sandgren A, Hollo V, van
der Werf MJ. Extrapul-
monary tuberculosis in
the European Union and
European Economic
Area, 2002 to 2011. Euro
Surveill. 2013;18.
[PMID: 23557943]
36. Schutz C, Meintjes G,
Almajid F, Wilkinson RJ,
Pozniak A. Clinical man-
agement of tuberculosis
and HIV-1 co-infection.
Eur Respir J. 2010;36:
1460-81. [PMID:
20947678]
37. Mendelson M. Diagnos-
ing tuberculosis in HIV-
infected patients: chal-
lenges and future
prospects. Br Med Bull.
2007;81-82:149-65.
[PMID: 17470475]
38. Lawn SD, Edwards DJ,
Kranzer K, Vogt M, Bek-
ker LG, Wood R. Urine
lipoarabinomannan
assay for tuberculosis
screening before antiret-
roviral therapy diagnostic
yield and association
with immune reconstitu-
tion disease. AIDS. 2009;
23:1875-80. [PMID:
20108382]
39. Kranzer K, Houben RM,
Glynn JR, Bekker LG,
Wood R, Lawn SD. Yield
of HIV-associated tuber-
culosis during intensified
case finding in resource-
limited settings: a sys-
tematic review and meta-
analysis. Lancet Infect
Dis. 2010;10:93-102.
[PMID: 20113978]
40. Malekmohammad M,
Marjani M, Tabarsi P,
Baghaei P, Sadr Z,
Naghan PA, et al. Diag-
nostic yield of post-
bronchoscopy sputum
smear in pulmonary
tuberculosis. Scand J
Infect Dis. 2012;44:369-
73. [PMID: 22497518]
www.ncbi.nlm.nih.gov
/pubmed/22497518 on
20 December 2016.
41. Zumla A, James DG.
Granulomatous infec-
tions: etiology and classi-
fication. Clin Infect Dis.
1996;23:146-58. [PMID:
8816144]
42. Chihota VN, Grant AD,
Fielding K, Ndibongo B,
van Zyl A, Muirhead D,
et al. Liquid vs. solid
culture for tuberculosis:
performance and cost in
a resource-constrained
setting. Int J Tuberc Lung
Dis. 2010;14:1024-31.
[PMID: 20626948]
ITC25 姝 2017 American College of Physicians
 drug resistance require a change When should clinicians refer
to a second-line regimen and have patients with suspected active
significantly worse outcomes than TB to an expert?
those with drug-susceptible TB (1). Providers who rarely care for pa-
Patients with MDR TB and resis- tients with TB should consult an
tance to 2 of the second-line expert if the patient has a negative
drugs, a fluoroquinolone and an AFB smear and culture results but
injectable aminoglycoside (e.g., pulmonary TB is still suspected
amikacin, capreomycin, or kana- clinically, if the patient has drug-
mycin), are considered to have resistant TB, or if the patient has
43. Update: nucleic acid
amplification tests for
extensively drug-resistant (XDR) TB been treated previously for TB.
tuberculosis. MMWR and have extremely poor treat- Delayed or inappropriate treat-
Morb Mortal Wkly Rep.
2000;49:593-4. Ac- ment outcomes, partially due to ment may lead to prolonged trans-
cessed at https://stacks the few good drugs left to treat mission of disease and develop-
.cdc.gov/view/cdc/27233
on 20 December 2016. them (48). ment of drug resistance.
44. Updated guidelines for
the use of nucleic acid
amplification tests in the
diagnosis of tuberculosis.
MMWR Morb Mortal Diagnosis... Patients should be tested for TB if they have prolonged
Wkly Rep. 2009;58:7-10. cough, hemoptysis, chest pain, or systemic symptoms. Clinical suspi-
[PMID: 19145221]
45. Steingart K, Schiller I,
cion should be higher in patients who are HIV-positive, even if they
Horne D, Boehme CC, have few symptoms and a clear chest radiograph. Patients with nega-
Dendukuri N. Xpert® tive microbiologic testing but for whom clinical suspicion is still high
MTB/RIF assay for pulmo-
nary tuberculosis and should receive empirical treatment with assessment for clinical re-
rifampicin resistance in sponse after 2 months. Patients should be referred to a TB expert if the
adults. Cochrane Data-
base Syst Rev. 2014;(1):
primary physician has little experience with TB, the presentation is not
CD009593. www.ncbi straightforward, drug resistance is present, or the patient does not re-
.nlm.nih.gov/pubmed spond to therapy.
/19145221 on 20
December 2016.
46. World Health Organiza-
tion. Policy statement:
automated real-time CLINICAL BOTTOM LINE
nucleic acid amplification
technology for rapid and
simultaneous detection
of tuberculosis and ri-
fampicin resistance:
Xpert MTB/RIF system.
Geneva: World Health
Organization; 2011.
Accessed at http://apps
Treatment
.who.int/iris/bitstream What is the standard drug should be given with INH to all
/10665/44586/1
/9789241501545_eng
treatment for active TB in persons at risk for neuropathy
.pdf on 20 December cases of drug-susceptible (i.e., pregnant women; breast-
2016.
47. Luetkemeyer AF, Firn- M tuberculosis infection? feeding infants; HIV-infected per-
haber C, Kendall MA, Wu
X, Mazurek GH, Benator The most recent professional sons; patients with diabetes, al-
DA, et al; AIDS Clinical
Trials Group A5295 and
guidelines for treating drug- coholism, malnutrition, or chronic
Tuberculosis Trials Con- susceptible TB were published in renal failure; and persons of ad-
sortium Study 34 Teams.
Evaluation of Xpert MTB/
June 2016 (49). The preferred vanced age). If given correctly
RIF versus AFB smear regimen for treating adults with and taken completely, this regi-
and culture to identify
pulmonary tuberculosis TB that is known or believed to men has a 95% cure rate (50).
in patients with sus- be drug-susceptible consists of
pected tuberculosis from The preferred schedule is daily
low and higher preva- an intensive phase of 4 drugs dosing during both the intensive
lence settings. Clin Infect
(INH, RIF, ethambutol [EMB], and
Dis. 2016;62:1081-8. and continuation phases with
[PMID: 26839383] pyrazinamide [PZA]) for 2
48. Migliori GB, Lange C, DOT 5 days per week (49). If
Girardi E, Centis R, Be- months, followed by a continua-
sozzi G, Kliiman K, et al; daily DOT is difficult, intermittent
SMIRA/TBNET Study
tion phase consisting of 2 drugs
Group. Extensively drug- (INH plus RIF), usually for 4 dosing can be done as an alter-
resistant tuberculosis is
months. If the M tuberculosis iso- native in HIV-negative, pulmo-
worse than multidrug-
resistant tuberculosis: late is found to be susceptible to nary TB cases caused by drug-
different methodology
and settings, same re- INH and RIF, EMB can be susceptible organisms without
sults [Letter]. Clin Infect stopped early in the intensive cavitation, but this should be re-
Dis. 2008;46:958-9.
[PMID: 18288911] phase. Pyridoxine (vitamin B6) viewed with a TB specialist.

姝 2017 American College of Physicians ITC26 In the Clinic Annals of Internal Medicine 7 February 2017

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 What is the best way to
monitor the results of
treatment and to detect adverse
effects during active TB
therapy?
Table 4 shows recommended
baseline, follow-up, and end-of-
treatment evaluations for patients
treated with first-line TB medica-
tions (49). The baseline visit
should assess AFB smear and TB
culture status as well as drug sus-
ceptibility of the infecting isolate
with rapid molecular diagnostic
and traditional susceptibility test-
ing. A chest radiograph should
be done to assess the extent of
cavitation. Blood should be
drawn to assess general health,
including liver and kidney func-
tion. Weight should be recorded,
and baseline symptoms should
be assessed. Patients should be
screened for HIV, hepatitis B and
C viruses, alcoholism, and diabe-
Table 4. Recommended Baseline, Follow-up, and End-of-Treatment
Evaluations for Patients Treated With First-Line TB Medications*
Test
Microbiology
Sputum smears and culture
Drug susceptibility testing
Imaging
Chest radiograph or other imaging
Clinical Assessment
Weight
Symptom and adherence review
Vision assessment
Laboratory Testing
AST, ALT, bilirubin, alkaline phosphate
Platelet count
Creatinine
HIV serology
Hepatitis B and C screen
Diabetes screen
ALT = alanine aminotransferase; AST = aspartate aminotransferase; TB = tuberculosis.
* Adapted from reference 49.
• = recommended diagnostics or procedures; ° = optional or contingent diagnostics or pro-
cedures.
7 February 2017 Annals of Internal Medicine
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
tes—these are common comor-
bidities among TB patients and
are associated with variable treat-
ment outcomes.
Sputum specimens should be
collected at treatment months 1
and 2 to assess response and,
beyond that point, if treatment
failure or delayed conversion is
suspected. If a patient is culture-
positive at treatment month 3,
drug-susceptibility testing should
be repeated to assess for ac-
quired drug resistance.
Medication adverse effects are
common. In a prospective cohort
study of Chinese patients receiv-
ing standard first-line treatment,
15% had a documented adverse
drug reaction that led to interrup-
tion or discontinuation of ther-
apy; of these, 7.7% resulted in
hospitalization, disability, or
death (51). The most serious ad-
Month of Treatment Completed
Baseline 1 2 3 4 5 6 7 8 End of
treatment
• • • ° ° °
• °
• • °
• • • • • • • • •
49. Nahid P, Dorman SE,
• • • • • • • • • Alipanah N, Barry PM,
• • • ° ° ° ° ° ° Brozek JL, Cattamanchi
A, et al. Official American
Thoracic Society/Centers
for Disease Control and
Prevention/Infectious
• ° ° ° ° ° ° ° ° Diseases Society of Amer-
ica clinical practice guide-
• ° ° ° ° ° ° ° ° lines: treatment of drug-
• ° ° ° ° ° ° ° ° susceptible tuberculosis.
Clin Infect Dis. 2016;63:
• e147-95. [PMID:
° 27516382]
50. Zumla A, Raviglione M,
° Hafner R, von Reyn CF.
Tuberculosis. N Engl J
Med. 2013;368:745-55.
[PMID: 23425167] www
.ncbi.nlm.nih.gov
/pubmed/23425167 on
20 December 2016.
In the Clinic ITC27 姝 2017 American College of Physicians
 51. Lv X, Tang S, Xia Y, Wang
X, Yuan Y, Hu D, et al.
Adverse reactions due to
directly observed treat-
ment strategy therapy in
Chinese tuberculosis
patients: a prospective
study. PLoS One. 2013;
8:e65037. [PMID:
23750225]
52. Saukkonen JJ, Powell K,
Jereb JA. Monitoring for
tuberculosis drug hepa-
totoxicity: moving from
opinion to evidence
[Editorial]. Am J Respir
Crit Care Med. 2012;
185:598-9. [PMID:
22422902]
53. Benator D, Bhattacharya
M, Bozeman L, Burman
W, Cantazaro A, Chaisson
R, et al; Tuberculosis
Trials Consortium. Rifap-
entine and isoniazid
once a week versus ri-
fampicin and isoniazid
twice a week for treat-
ment of drug-susceptible
pulmonary tuberculosis
in HIV-negative patients:
a randomised clinical
trial. Lancet. 2002;360:
528-34. [PMID:
12241657]
54. Jo KW, Yoo JW, Hong Y,
Lee JS, Lee SD, Kim WS,
et al. Risk factors for
1-year relapse of pulmo-
nary tuberculosis treated
with a 6-month daily
regimen. Respir Med.
2014;108:654-9. [PMID:
24518046]
55. Horne DJ, Royce SE,
Gooze L, Narita M,
Hopewell PC, Nahid P,
et al. Sputum monitoring
during tuberculosis treat-
ment for predicting out-
come: systematic review
and meta-analysis. Lan-
cet Infect Dis. 2010;10:
387-94. [PMID:
20510279]
56. Chang KC, Leung CC,
Yew WW, Ho SC, Tam
CM. A nested case-
control study on
treatment-related risk
factors for early relapse
of tuberculosis. Am J
Respir Crit Care Med.
2004;170:1124-30.
[PMID: 15374844]
57. Availability of an assay
for detecting Mycobacte-
rium tuberculosis, includ-
ing rifampin-resistant
strains, and consider-
ations for its use—United
States, 2013. MMWR
Morb Mortal Wkly Rep.
2013;62:821-7. [PMID:
24141407] Accessed at
www.ncbi.nlm.nih.gov
/pubmed/24141407 on
20 December 2016.
58. Thwaites G, Fisher M,
Hemingway C, Scott G,
Solomon T, Innes J;
British Infection Society.
British Infection Society
guidelines for the diag-
nosis and treatment of
tuberculosis of the cen-
tral nervous system in
adults and children. J
Infect. 2009;59:167-87.
[PMID: 19643501]
姝 2017 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
verse effect is often hepatotoxic-
ity, which occurred in 3%–13% of
patients in a review of retrospec-
tive studies (52). Patients are en-
couraged to abstain from alcohol
and avoid acetaminophen-
containing products while receiv-
ing TB treatment. Persons who
have evidence of hepatotoxicity
at baseline, develop symptoms
of hepatotoxicity, chronically
consume alcohol, have viral hep-
atitis or a history of liver disease,
or are HIV co-infected should
have liver function tests regularly
throughout treatment. Patients
on an EMB-containing regimen
should have their vision assessed
at baseline and on a monthly ba-
sis thereafter with a visual acuity
test (e.g., Snellen test) and a
color discrimination test (e.g.,
Ishihara test).
When and why should the
standard treatment regimen be
modified?
During treatment, a sputum
specimen for AFB and mycobac-
terial culture should be obtained
monthly until results are negative
for 2 consecutive months. Posi-
tive culture results after 2 months
of therapy have been shown to
predict relapse after therapy
completion, although sensitivity
is low (53–55). Cavitation on ini-
tial chest radiograph has also
been shown to predict relapse
(53, 56). In patients treated for 6
months, a positive culture at 2
months and cavitation were asso-
ciated with a relapse rate of al-
most 20% within 1 year of treat-
ment completion compared with
2% in patients with neither factor
(54). Expert opinion based on
these findings recommends ex-
tending the continuation phase
for an additional 3 months (i.e., a
continuation phase of 7 months
leading to a total of 9 months'
treatment) (49).
Because of the complexity,
length, and adverse effects of
treatment, complete adherence
ITC28 In the Clinic
is challenging. Breaks earlier in
treatment and of longer duration
can have more serious effects
and may warrant restarting the
treatment “clock” (49). Specifi-
cally, if treatment is interrupted
for 2 weeks or more during the
intensive phase, the patient
should restart this phase from the
beginning. If treatment is inter-
rupted for less than 2 weeks in
the intensive phase, continuation
of therapy until all doses are
completed is acceptable as long
as this occurs within 3 months.
For the continuation phase, if a
patient takes more than 80% of
all doses and AFB smear results
were initially negative, further
therapy may not be necessary. If
the AFB was positive, the patient
should take all doses until com-
pleted. If more than 80% of all
doses have been missed but dur-
ing less than 2 consecutive
months, the patient can continue
therapy until completion. If more
than 2 consecutive months were
missed, the patient should repeat
a full course of therapy from the
beginning (intensive and contin-
uation phases) (57).
How should patients with
extrapulmonary TB be treated?
The principles for treating pul-
monary TB apply to extrapulmo-
nary disease. Increasing evi-
dence suggests that 6 –9 months
of INH- and RIF-containing regi-
mens are effective for TB treat-
ment in most extrapulmonary
disease sites, although the ro-
bustness of the data for extrapul-
monary disease is not as great as
for pulmonary TB (49). The clear
exception is TB meningitis, for
which the ideal treatment regi-
men is not known but most ex-
perts recommend 2 months of
the 4-drug regimen and then 10
months of INH plus RIF (58). A
mortality benefit has also been
shown with the addition of corti-
costeroids (dexamethasone or
Annals of Internal Medicine 7 February 2017
 prednisolone) to this regimen,
tapered over 6 – 8 weeks (59, 60).
No large randomized, controlled
trials or systematic reviews found
differences in mortality, cardiac
tamponade, or constrictive peri-
carditis among patients with TB
pericarditis who did or did not
receive steroids (49, 61). Regard-
less, experts recommend that
addition of steroids be consid-
ered in patients at highest risk for
inflammatory complications (i.e.,
presence of large pericardial ef-
fusions, high levels of inflamma-
tory cells or markers in pericar-
dial fluid, or early signs of
pericardial constriction).
For monitoring treatment re-
sponse, bacteriologic evaluation
is often limited by difficulty in ob-
taining follow-up specimens.
Sputum specimens should be
obtained when concurrent pul-
monary TB is present; otherwise,
response to therapy is often
judged on clinical and radio-
graphic findings.
What is the approach to
treatment of active TB and HIV
co-infection?
Because patients with HIV infec-
tion often have more rapid pro-
gression to TB and higher mortal-
ity related to delayed treatment
initiation, TB medication should
be started as soon after diagno-
sis as possible (62). The recom-
mended regimen for drug-
susceptible TB is the same for
co-infected patients and HIV-
uninfected patients. For HIV-
infected patients receiving anti-
retroviral therapy (ART) (which
should be all patients), the stan-
dard 6-month daily regimen for
drug-susceptible TB is recom-
mended (49). In rare cases where
an HIV-infected person does not
receive ART during TB treatment,
the continuation phase can be
extended for an additional 3
months (i.e., a total of 9 months'
treatment). As with HIV-
7 February 2017 Annals of Internal Medicine
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
uninfected patients, regimens
should be extended for the
reasons outlined above. HIV-
infected patients should not
receive intermittent dosing regi-
mens, regardless of whether they
are receiving ART, because these
regimens have been associated
with higher rates of relapse and
acquired drug resistance in co-
infected persons (53, 63).
Management of patients co-
infected with HIV and TB is com-
plicated by the timing of ART ini-
tiation, potential drug
interactions between HIV and TB
medications, and paradoxical
reactions (i.e., immune reconsti-
tution inflammatory syndrome
[IRIS] [see below]).
Research now strongly supports
early initiation of ART for patients
with TB. A systematic review
found that the overall reduction
in mortality with ART initiation
during TB treatment was 24%
(relative risk [RR], 0.76 [95% CI,
0.57–1.01]) (49). Overall risk for
HIV disease progression was re-
duced by 34% by early or imme-
diate ART (RR, 1.88 [CI, 1.31–
2.69]). Based on this review,
patients with a CD4 count less
than 0.050 × 109 cells/L should
be started on ART within 2 weeks
after initiation of TB therapy, and
those with a CD4 count of
0.050 × 109 cells/L or greater
should be started between 8 and
12 weeks after initiation. The im-
portant exception is HIV-infected
patients with TB meningitis, in
whom ART should not be initi-
ated during the first 8 weeks of
anti-TB therapy at any CD4 count
to avoid increased morbidity
from IRIS.
The major concern with regard to
drug interactions in co-infected
patients is that between RIF, a
potent inducer of multiple drug-
metabolizing enzymes and drug
transporters, and ART. Although
RIF has been shown to decrease
In the Clinic
59. Critchley JA, Young F,
Orton L, Garner P. Corti-
costeroids for prevention
of mortality in people
with tuberculosis: a sys-
tematic review and meta-
analysis. Lancet Infect
Dis. 2013;13:223-37.
[PMID: 23369413]
60. Thwaites GE, Nguyen DB,
Nguyen HD, Hoang TQ,
Do TTO, Nguyen TCT,
et al. Dexamethasone for
the treatment of tubercu-
lous meningitis in ado-
lescents and adults.
N Engl J Med. 2004 Oct
21;351(17):1741–51.
[PMID: 15496623] Ac-
cessed at www.ncbi.nlm
.nih.gov/pubmed
/15496623 on 20
December 2016.
61. Mayosi B, Ntsekhe M,
Smieja M. Immunother-
apy for tuberculous peri-
carditis. N Engl J Med.
2014;371:2532-3.
[PMID: 25539114]
62. Holtz TH, Kabera G, Mthi-
yane T, Zingoni T, Nade-
san S, Ross D, et al. Use
of a WHO-recommended
algorithm to reduce
mortality in seriously ill
patients with HIV infec-
tion and smear-negative
pulmonary tuberculosis
in South Africa: an obser-
vational cohort study.
Lancet Infect Dis. 2011;
11:533-40. [PMID:
21514234]
63. Vernon A, Burman W,
Benator D, Khan A, Boze-
man L. Acquired rifamy-
cin monoresistance in
patients with HIV-related
tuberculosis treated with
once-weekly rifapentine
and isoniazid. Tuberculo-
sis Trials Consortium.
Lancet. 1999 May 29;
353:1843-7. [PMID:
10359410]
ITC29 姝 2017 American College of Physicians

64. Tuberculosis (TB): Man-
aging drug interactions
in the treatment of HIV-
related tuberculosis.
2013. Accessed at www
.cdc.gov/tb/publications
/guidelines/tb-
_hiv_drugs/table1a.htm
on 21 December 2016.
65. Török ME, Yen NT, Chau
TT, Mai NT, Phu NH, Mai
PP, et al. Timing of initia-
tion of antiretroviral
therapy in human immu-
nodeficiency virus (HIV)—
associated tuberculous
meningitis. Clin Infect
Dis. 2011;52:1374-83.
[PMID: 21596680]
66. Horsburgh CR, Barry CE,
Lange C. Treatment of
tuberculosis. N Engl J
Med. 2015 Nov 26;373:
2149-60. [PMID:
26605929]
67. World Health Organiza-
tion. WHO treatment
guidelines for drug-
resistant tuberculosis—
2016 update. Geneva:
World Health Organiza-
tion; 2016.
姝 2017 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017
drug levels of efavirenz,
efavirenz-based ART with RIF-
containing TB treatment is the
most well-characterized regimen
with the best-documented out-
comes for patients and low dis-
continuation rates; as such, it
remains the preferred drug com-
bination (64). Alternative regi-
mens that include protease inhibi-
tor– based ART paired with a
rifabutin-containing TB regimen
and nevirapine-based ART paired
with a RIF-containing TB regimen
have also been endorsed if a pa-
tient cannot take efavirenz. If possi-
ble, co-infected patients should be
managed by providers with expe-
rience treating the 2 diseases due
to the complexity of the treatment
regimens.
Co-infected patients are at risk
for paradoxical worsening of TB
symptoms and lesions after be-
ginning TB therapy and ART. This
reaction is a consequence of re-
constitution of the immune re-
sponse brought on by ART and is
referred to as IRIS. IRIS in patients
with TB is reported more com-
monly in those with CD4 counts
less than 0.050 × 109 cells/L. Be-
fore it can be diagnosed, other
reasons for the worsening signs
and symptoms need to be ex-
cluded, especially treatment fail-
ure due to drug resistance or
other opportunistic diseases.
Patients who start ART within 2
weeks of TB therapy have higher
IRIS rates than those who start
between 8 and 12 weeks; hence,
the recommendation to delay
ART in patients with higher CD4
counts. In patients with a CD4
cell count less than 50 per cubic
millimeter, starting ART within 2
weeks after TB treatment initia-
tion is recommended. Most IRIS
can be managed symptomati-
cally by adding anti-inflammatory
agents, such as ibuprofen or cor-
ticosteroids (49). The exception is
central nervous system TB, in-
cluding TB meningitis: Patients in
ITC30 In the Clinic
whom ART was initiated within
2 weeks of TB therapy had in-
creased rates of adverse events
and higher mortality (65).
What are the indications for
DOT for active TB?
Adherence to TB treatment is chal-
lenging but critical for both im-
proved patient outcomes and to
avoid disease transmission; there-
fore, DOT (in which ingestion of
each dose is witnessed) has been
widely endorsed. DOT remains the
standard of practice in most U.S.
and European TB programs (49)
and is discussed in detail else-
where (49).
When should patients be
treated for drug-resistant TB,
and what is the basic
approach?
Patients infected with M tubercu-
losis that is resistant only to INH
(i.e., INH-monoresistant TB) can
receive the 3 other first-line
drugs (RIF, EMB, and PZA) for the
full 6 months of treatment, or a
fluoroquinolone (levofloxacin or
moxifloxacin) could be added to
the 3-drug regimen (66). Patients
with MDR TB (i.e., documented
resistance to both INH and RIF),
or RIF-resistant TB need a greater
adjustment. The WHO has rec-
ommended, on the basis of a
large retrospective meta-analysis,
that patients with MDR TB re-
ceive 5 drugs to which their iso-
late is susceptible in the induc-
tion phase, which should last 6 to
8 months (67). This regimen ide-
ally should include a fluoroquino-
lone and an injectable aminogly-
coside (amikacin, kanamycin, or
capreomycin). Patients should
then enter a continuation phase
with 4 of those drugs (with re-
moval of the injectable) for an
additional 12–18 months, result-
ing in a total of 18 –24 months of
treatment. Ideally, these regi-
mens should be tailored to the
susceptibility profile of the spe-
cific patient's infection.
Annals of Internal Medicine 7 February 2017
 In May 2016, the WHO endorsed
the “shorter” MDR TB regimen,
previously known as the “Bangla-
desh regimen.” It consists of 7
drugs given for 9 –12 months,
with cure rates of 85%– 89% and
few relapses in specific settings.
Clinical trials in additional set-
tings are under way, but the
WHO encourages use of this reg-
imen in MDR TB patients who 68. World Health Organiza-
tion. The Shorter MDR-TB
have no evidence of resistance to Regimen. Geneva: World
Health Organization;
any of the 7 drugs, have no his- 2016. Accessed at www
tory of exposure to second-line .who.int/tb/Short_MDR
_regimen_factsheet.pdf
TB medications for more than a on 20 December 2016.
month or intolerance to any of
the medications and low risk for
toxicity, are not pregnant, and do
not have extrapulmonary disease
(68).

Treatment... Active TB should be treated with a regimen and for a

length of time tailored to patient characteristics and organism suscepti-
bility. Patients with complications or adverse drug effects that cannot be
managed at home should be hospitalized, but patients should not be
hospitalized solely for isolation. Patients should be monitored at least
monthly for clinical response and adverse drug effects. Sputum of pa-
tients whose initial cultures are positive should be cultured monthly un-
til negative. DOT should be considered, especially for patients who do
not respond to therapy, have HIV infection, are receiving intermittent
drug regimens, or have drug-resistant TB. Patients who have HIV co-
infection, are drug-resistant, are failing therapy, or have extrapulmonary
disease ideally should be managed in consultation with experts.

CLINICAL BOTTOM LINE

In the Clinic Patient Information

IntheClinic
https://medlineplus.gov/tuberculosis.html

Tool Kit
Information on tuberculosis from the National Institutes
of Health MedlinePlus.
www.mayoclinic.org/diseases-conditions/tuberculosis
/home/ovc-20188556
Information on tuberculosis from the Mayo Clinic that is
useful to both patients and medical professionals.
Tuberculosis www.cdc.gov/tb/publications/pamphlets/TBgtfctsEng.pdf
Patient handout on tuberculosis from the Centers for
Disease Control and Prevention.
www.cdc.gov/tb/esp/publications/pamphlets/TBgtf-
ctsSpan.PDF
Patient handout in Spanish from the Centers for Disease
Control and Prevention.

Guidelines
www.cdc.gov/tb/publications/guidelines/default.htm
Access guidelines on a variety of specific topics related to
tuberculosis.
www.who.int/publications/guidelines/tuberculosis/en
World Health Organization guidelines on tuberculosis.

Other Information
www.who.int/tb/en
The World Health Organization provides information
about global efforts to eliminate tuberculosis and re-
lated programs.

7 February 2017 Annals of Internal Medicine ITC31 姝 2017 American College of Physicians

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017

 WHAT YOU SHOULD In the Clinic
Annals of Internal Medicine
KNOW ABOUT
TUBERCULOSIS
What Is Tuberculosis?
Tuberculosis (TB) is a disease caused by bacteria
that attack the lungs. There are 2 kinds of TB:
• Active TB makes a person feel sick and can spread
to others by coughing, sneezing, or speaking.
• Latent TB does not make you feel sick and
does not spread to others. People with latent
TB are infected with the bacteria but do not
have the disease. Some people with latent TB
may progress to active TB later on. People
who are older, have HIV infection or other
immunosuppression, have poor nutrition, or
other health problems are at greater risk.
What Are the Warning Signs?
Symptoms of active TB in the lungs include the
following: How Is It Treated?
• A long-lasting cough • Active TB can be treated and usually cured by
• Chest pain taking medicines for 6 or more months.
• Coughing up blood or mucus • Latent TB can be treated with medicine taken
• Weight loss for 3 to 9 months. This medicine will help
• Feeling weak or tired prevent you from getting active TB.
• Fever and chills • If TB is not treated, it can be deadly.
• Night sweats
Latent TB has no symptoms. What Is Drug-Resistant TB?
• Drug-resistant TB happens when the
How Is It Diagnosed? medicines usually used to treat TB do not
There are 2 types of tests for TB infection: a skin work. The bacteria causing the disease is
“resistant” to the medicines.

test and a blood test. These tests can check for
both latent and active TB but cannot tell you
which type you have. If either test is positive,
your doctor will test your sputum and order a
chest X-ray. This will help him or her know
whether you have active TB that can spread to
other people. People should be tested for TB
infection if they:
• Have had close contact with someone who has
active TB
• Have certain chronic illnesses, such as HIV
• Have come to the United States in the past 5
years from a country where TB is common
• Work with bacteria in laboratories
• Are health care workers
• Live or work in facilities where many people
congregate, such as jails, nursing homes, and
homeless shelters
• Are smokers or drug users
Patient Information
• This type of TB is rare in the United States but
is becoming more common in other parts of
the world.
• It is usually treated with more and stronger
medicines for a long period.
Questions for My Doctor
• If I have latent TB, is it safe for me to be
around other people?
• How did I get TB?
• Will I develop active TB?
• Do I need treatment?
• What treatment is best for me?
• How long do I have to stay on treatment?
• What medicines are safe to take with TB
medicines?
• How long will it take to be cured?
• Can I leave my house and go to work?

For More Information

MedlinePlus
https://medlineplus.gov/tuberculosis.html
Centers for Disease Control and Prevention
www.cdc.gov/tb/faqs

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/936047/ by a NYU Medical Center Library User on 04/21/2017