C u r re n t M e d i c a l

Management of Pulmonary
Tub e rc u l o s i s
Robert W. Belknap, MDa,b,*

KEYWORDS
 Pulmonary tuberculosis  TB diagnosis  TB treatment  Drug-susceptible TB  Drug-resistant TB

KEY POINTS
 Diagnosing pulmonary tuberculosis early requires recognizing the various symptoms and radio-
graphic presentations of disease.
 Nucleic acid amplification tests are more sensitive and specific than sputum smears and are
increasingly used.
 Isoniazid, rifampin, pyrazinamide, and ethambutol remain the first-line treatment of pulmonary
tuberculosis worldwide.
 Newer and repurposed medications offer the potential to improve treatment outcomes for drug-
resistant tuberculosis (eg, bedaquiline, delamanid, pretomanid, linezolid, moxifloxacin, levofloxa-
cin, and clofazimine).
 Treatment of patients with comorbidities like HIV, renal failure, and liver disease is more compli-
cated but can be managed.

INTRODUCTION more common in young children and people with

immunosuppression. For most, their immune sys-
Tuberculosis (TB) has afflicted people for millennia tem controls the infection, ultimately forming gran-
and, unfortunately, despite effective treatment ulomas without ever being symptomatic.
was the leading cause of death from an infection TB can remain hidden in individuals for decades
in 2016, killing 1.7 million people globally.1,2 TB is before progressing to cause clinical disease. Reac-
also a common cause of disease and death among tivation may be associated with other conditions
people living with HIV. Recent estimates suggest that weaken the immune system, such as HIV or dia-
that one-quarter of the world’s population is betes, but often occurs in otherwise healthy people.
infected with TB and more than 10 million people When it occurs, reactivation TB involves the lungs in
develop active disease each year.1,3 Infection oc- approximately 75% of people.1,4 Those with pulmo-
curs primarily through airborne transmission and nary TB are then able to spread it to others, perpet-
usually requires prolonged exposure to a person uating the cycle of infection that leads to disease.
with pulmonary TB. The initial infection results in Early diagnosis and treatment of pulmonary TB
spread of organisms from the lungs to local lymph help minimize TB transmission and are an important
nodes followed by dissemination through the strategy for decreasing the global TB burden. Un-
blood. Approximately 5% of people develop pro- treated, smear-positive pulmonary TB has a
gressive disease after initial infection and this is
thoracic.theclinics.com

Disclosure: The author has nothing to disclose.

a
Denver Metro Tuberculosis Program, Denver Health and Hospital Authority, 605 Bannock, Denver, CO 80204,
USA; b Division of Infectious Diseases, Department of Medicine, University of Colorado Denver Anschutz Med-
ical Campus, 12700 East 18th Avenue, RC2, Aurora, CO 80045, USA
* Denver Metro Tuberculosis Program, 605 Bannock, Denver, CO 80204.
E-mail address: robert.belknap@dhha.org

Thorac Surg Clin 29 (2019) 27–35

https://doi.org/10.1016/j.thorsurg.2018.09.004
1547-4127/19/Ó 2018 Elsevier Inc. All rights reserved.
28 Belknap

10-year mortality of 70%.5 Late diagnoses also chest radiograph is generally sufficient for evalu-
result in permanent lung damage for many patients.6 ating adults and adolescents. A lateral chest radio-
graph is helpful in younger children to look for hilar
DIAGNOSIS adenopathy and retrocardiac infiltrates. Classi-
Risk and Symptoms cally described radiographic findings are upper
lobe fibronodular opacities with or without cavita-
TB disease first requires infection, which usually oc- tion (Fig. 1). Miliary TB appears as small nodules
curs after prolonged exposure to someone with scattered diffusely throughout both lungs (Fig. 2).
active pulmonary TB. Known close contact to a TB can present with consolidations that look
person with pulmonary TB is the greatest risk but radiographically similar to bacterial pneumonia.
is uncommon. Most people do not know when Empiric antibiotics for bacterial infection can
they were exposed and infected so being born or cause a delayed TB diagnosis. This is particularly
living in a TB-endemic country is the most common true when fluoroquinolones are prescribed
risk. Currently that includes most countries, with the because they have excellent activity against
exceptions of Western and Northern Europe, the most strains. Other radiographic presentations of
United States, Canada, Australia, New Zealand, pulmonary TB include solitary or multiple nodules,
and Japan. Even within those countries, individual masses, adenopathy (Fig. 3), and pleural effusion.
risk may vary by location and demographic factors, CT scans of the chest show the extent of dis-
such as homelessness. ease in greater detail than plain radiographs but
The first step in diagnosing pulmonary TB re- are rarely needed for the diagnosis or manage-
quires having suspicion for disease. Delays in ment of TB. When pulmonary TB is suspected,
diagnosis are common because the symptoms CT scans should be deferred while collecting res-
are nonspecific. Unlike most infectious diseases piratory specimens for acid-fast bacilli (AFB). CT
that either resolve or get progressively worse, TB scans may be beneficial when respiratory speci-
may wax and wane even without treatment. As a mens are initially negative or other diagnoses are
result, patients may not recognize their symptoms considered likely. The chest CT findings in patients
as the same illness, resulting in delays presenting with active TB vary but may include tree-in-bud or
for care. TB should be considered in patients ground-glass opacities, fibrosis, nodules, cavita-
who present with a subacute illness, particularly tion, pleural effusions, and hilar or mediastinal
if they have lived in a country where TB is endemic. adenopathy.8
People with risk for TB infection and who have PET scans have been used in research studies to
certain other medical conditions are at higher risk better understand the pathophysiology of TB.9,10
for developing disease. Those at greatest risk are PET scans have also been suggested as a tool for
people living with HIV, adult and child contacts diagnosing TB and monitoring response to ther-
of pulmonary TB, patients initiating anti–tumor ne- apy.11 They are not able to differentiate TB from
crosis factor alpha treatment, patients receiving malignancy very well but could be helpful in
dialysis, patients receiving organ or hematologic
transplantation, and patients with silicosis. People
with diabetes are also at higher risk than the gen-
eral public, and diabetes is one of the most com-
mon comorbidities associated with active TB.7
Importantly, most people who develop active TB
do not have one of these associated conditions.
Typical symptoms for pulmonary TB include

 Cough greater than 3 weeks that is typically

productive of sputum, worsens over time,
and may be associated with hemoptysis
 Dyspnea
 Fever
 Night sweats
 Weight loss
 Chest pain (with pleural disease)

Imaging
Chest imaging should be done for anyone sus-
pected of having pulmonary TB. A posteroanterior Fig. 1. Right upper lobe fibrotic opacities.
 Management of Pulmonary Tuberculosis 29

Fig. 2. (A) Miliary TB on chest radiograph. (B) Miliary TB on CT.

identifying active lesions for biopsy. Cost and ac- for patients who are unable to expectorate
cess are likely to remain major barriers to routine sputum. Bronchoscopy should be considered for
use. patients who fail sputum induction or when an
alternative diagnosis is considered likely. In pa-
tients with miliary disease who undergo bronchos-
Microbiologic Specimens
copy, bronchial brushings or transbronchial
Sputum samples are the hallmark for diagnosing biopsy are more sensitive than bronchoalveolar
pulmonary TB and AFB smears remain the initial lavage.12
test. The sensitivity of AFB smears increases Nucleic acid amplification tests (NAATs) are
with serial specimens from approximately 54% more sensitive and specific than AFB smears for
with 1 specimen, 65% with 2 specimens, and up diagnosing TB. The GeneXpert (Cepheid, USA) re-
to 70% with 3 specimens.12 The quality of the quires minimal specimen processing and labora-
sputum is generally more important than the timing tory expertise that have made it feasible to use in
although an early morning sputa is more sensitive resource limited settings. A single GeneXpert can
than a single spot specimen.13 Early morning detect 97% of sputum smear positive TB and
sputa may also be easier to collect for patients nearly 60% of smear-negative, culture-positive
with minimal symptoms. Sputum collection every disease whereas a second test increases the
6 hours to 8 hours, particularly in hospitalized pa- sensitivity to 100% and 70%, respectively.15
tients, facilitates more rapid diagnosis and mini- Based on these results, GeneXpert was approved
mizes the time in isolation for people without TB. in the United States for removing hospitalized pa-
Bronchoscopy is not generally needed to diag- tients with negative results from respiratory isola-
nose pulmonary TB.14 Sputum induction is as tion. Many NAATs are also able to detect
effective, less expensive than, and less invasive resistance mutations, allowing a shorter time to

Fig. 3. (A) Paratracheal adenopathy (arrow) in a patient with HIV. (B) Resolution of adenopathy with treatment.
30 Belknap

starting effective treatment. Susceptibility and laboratory tests that are important include a com-
resistance should still be confirmed with culture plete blood cell count, hepatic function panel, and
whenever possible. creatinine. Tests done within the prior few months
Pleural TB typically presents with a unilateral, are usually sufficient. Baseline visual acuity and
exudative effusion that is lymphocyte predomi- color vision should be documented and monitored
nant. AFB smears and cultures from pleural fluid monthly for people receiving ethambutol (EMB).
generally detect less than 50% of active TB.12 Additional testing and monitoring may be needed
Pleural biopsy with tissue sent for pathology and in patients with drug-resistant TB depending on
cultures is the best method for definitively diag- the medications used.
nosing pleural TB. Closed needle biopsy is effec-
tive in most patients whereas thoracoscopic TREATMENT
biopsy for histopathology and culture is diagnostic Drug-Susceptible Tuberculosis
in nearly 100% of cases. Pleural fluid adenosine
Standard first-line treatment of pulmonary TB con-
deaminase has moderate sensitivity and speci-
sists of isoniazid (INH), rifampin (RIF),
ficity and may be helpful in settings where pleural
biopsy is not feasible.16
Culture remains the gold standard for diag- Table 1
Medications for drug-susceptible tuberculosis
nosing TB and should be done whenever possible.
Growing TB allows for a definitive diagnosis and Common or Severe Adverse
testing for drug susceptibility. Unfortunately, cul- Drug Effects
tures are rarely done for patients with newly diag-
nosed TB in resource limited settings. The cost INH  Headache
 Fatigue
and complexity of maintaining laboratory capacity
 Nausea/anorexia/abdominal
have prevented widespread use so cultures are pain
generally limited to those with treatment failure or  Drug-induced hepatitis
relapse. Liquid cultures provide the shortest time  Rash
to growth and first-line susceptibility testing. Solid  Peripheral neuropathy
media are beneficial for isolating TB when there is RIF  Drug-drug interactions
contamination and testing susceptibility to  Red/orange discoloration of
second-line drugs when needed. body fluids
 Rash
Tuberculin Skin Test and Interferon-Gamma  Nausea/anorexia/abdominal
Release Assays pain
 Flulike illness/hypersensitivity
The tuberculin skin test and interferon-gamma  Drug-induced hepatitis
release assays (IGRAs) are immunologic tests for  Acute renal failure
diagnosing TB infection. The 2 commercially avail-  Anemia/thrombocytopenia
able IGRAs are QuantiFERON-TB (Qiagen, Ger- PZA  Nausea/anorexia/abdominal
many) and T-SPOT.TB (Oxford Immunotec, pain
United Kingdom). These tests should never be  Drug-induced hepatitis
used to rule out TB and have a limited role in diag-  Rash/acute flushing
nosing active disease. The tuberculin skin test  Join pain
misses up to 30% of people with active TB  Elevated uric acid
whereas the IGRAs miss approximately 10% to EMB  Visual impairment/optic
15%.17 They may be helpful when trying to decide neuritis
whether to treat someone empirically for TB who Levofloxacin  Headache/insomnia
otherwise has negative AFB smears or NAATs  Nausea/anorexia/abdominal
while waiting for cultures. These decisions are pain
 Rash
best made in consultation with the local public
 QT prolongation
health providers who can evaluate the public  Tendonitis/tendon rupture
health risk of delaying treatment.
Moxofloxacin  Headache/insomnia
 Nausea/anorexia/abdominal
Other Testing pain
Additional testing is important for the management  Rash
of patients with suspected or confirmed TB but  QT prolongation
 Tendonitis/tendon rupture
does not aid in the diagnosis. All patients with sus-
 Drug-induced hepatitis
pected TB should be tested for HIV. Other
 Management of Pulmonary Tuberculosis 31

pyrazinamide (PZA), and EMB.18,19 Table 1 lists 4 months was as effective as the standard
the common and potential severe adverse effects regimen. Potential adverse effects from levofloxa-
from these medications. With this combination, cin and moxifloxacin are listed in Table 1.
INH has the best early bactericidal activity
whereas RIF is necessary to cure TB in the short- Multidrug-Resistant and Extensively Drug-
est time. PZA is effective against minimally active Resistant Tuberculosis
organisms and inclusion in the first 2 months facil- Multidrug-resistant (MDR) TB is defined as resis-
itates cure with a 6-month course. EMB is the least tance to INH and RIF with or without resistance to
active in this combination but is protective against other drugs. Empiric treatment should include 4 to
acquired resistance when treatment is started 6 drugs that are likely to be active.25,26 Fluoroquino-
without knowing the susceptibilities. This combi- lones and injectable drugs have been the corner-
nation of medications is given for 2 months in the stone of treatment of MDR TB. Moxifloxacin is
initial phase of TB treatment. With fully susceptible often chosen over levofloxacin based on in vitro
TB, EMB is not needed and the initial 2 months of data but both drugs are effective.27 All oral regi-
treatment can be with INH, RIF, and PZA alone. mens using newer drugs are being evaluated for ef-
The continuation phase for drug-susceptible TB ficacy and safety and may replace older regimens
is INH and RIF. The usual duration is 4 months to that include an injectable drug for many
complete 6 months of total treatment. There is months.28–30 Treatment durations had typically
no test to determine when a person has been been 18 months to 24 months but a 9-month
cured, so the goal of treatment is to achieve a regimen has shown promise and shorter durations
low risk for relapse (generally <5%). Factors using novel combinations are being evaluated.31–33
known to be associated with higher relapse rates Extensively drug-resistant (XDR) TB is resistant
are cavitation, positive sputum culture after to INH and RIF plus fluoroquinolones and at least
2 months of treatment, and being underweight at 1 second-line injectable drug (amikacin, kana-
the start and failing to gain weight during treat- mycin, or capreomycin).19,26 The basic principles
ment.20,21 A common approach is to extend treat- are the same as treating MDR TB. Treatment of
ment by 3 months (9 months total) for patients with XDR TB should be done in consultation with an
1 or more risks for relapse. Similarly, patients expert in treating drug-resistant TB. A general prin-
whose TB is resistant to PZA or who are cannot ciple is to include 5 to 6 drugs likely to have activ-
complete 2 months of that medication can be ity. PZA may be included despite phenotypic
treated with a 9-month course of INH and RIF.18 resistance with the hope that it might retain some
activity against the minimally active organisms.
Isoniazid Resistance or Intolerance Some providers also use high doses of moxifloxa-
cin and/or INH when low-level resistance to these
When INH cannot be given because of resistance drugs is found or when there are few other options.
or intolerance, a fluoroquinolone (moxifloxacin or Linezolid has become an important component of
levofloxacin) is often given in combination with treatment regimens for both MDR TB and XDR
RIF, PZA, and EMB for the initial phase. For the TB.34,35 Although the optimal dose is unknown,
continuation phase, the options are to continue observational studies giving one-half the usual
treatment with RIF, PZA, or EMB or to continue daily dose for bacterial infections seems effective
the fluoroquinolone and RIF with or without and has less toxicity. Bedaquiline is another newer
EMB.18 The combination of RIF, PZA, and EMB drug that is becoming increasingly important for
throughout is as effective as treatment with INH treating drug-resistant TB.36 Other new and repur-
and RIF for drug-susceptible TB but may not be posed medications that are showing efficacy
well tolerated.22 A retrospective evaluation of pa- include delamanid, pretomanid, and clofazimine.30
tients with INH-resistant TB found that patients
who received a fluoroquinolone in combination MONITORING TREATMENT
with RIF and EMB had good outcomes.23 A recent
study evaluated the role of moxifloxacin in different Ensuring adherence to treatment is important to
experimental arms with a primary goal of short- minimize the risk of failure, acquired drug resis-
ening treatment to 4 months. One experimental tance, and TB transmission. Directly observed
arm received moxifloxacin, RIF, PZA, and EMB therapy (DOT) maximizes treatment completion,
daily for 2 months followed by once-weekly moxi- allows close monitoring for drug-related side ef-
floxacin and rifapentine for 4 months.24 The 4- fects, and is the standard of care for patients
month treatment arms were inferior to the stan- with pulmonary TB. Intermittent dosing strategies
dard 6-month daily regimen but the experimental were developed to make DOT easier for patients
arm with the once-weekly continuation phase for and providers. Although outcomes with
32 Belknap

intermittent dosing were generally comparable to The timing of HIV treatment in people with pul-
daily therapy, recent meta-analyses have shown monary TB has been studied in several random-
higher rates of relapse including in patients with ized trials. These studies showed that survival
HIV.37–39 Therefore, current guidelines recom- was higher in people with CD4 counts less than
mend daily therapy unless intermittent therapy is 50 cells/mL who started ART within 2 weeks of
the only feasible option.18,19 Because daily DOT TB treatment. The exception is people with
is difficult and costly for patients and programs, concomitant central nervous system disease
digital technology is increasingly used as a more because they are at high risk for immune reconsti-
affordable and patient-centered approach to treat- tution inflammatory syndrome (IRIS). The optimal
ment monitoring.40–42 Some technologies allow timing to start ART in these individuals is unclear.47
patients and health care personnel to interact in Patients with CD4 counts greater than 50 cells/mL
real time.40,41 Others allow patients to record a can delay ART if needed but ideally should start
video of themselves taking the medications at a within the first 2 months. The highest risk for IRIS
time they prefer. The video is then uploaded to a is in the first 4 weeks to 12 weeks after starting
secure server and can be viewed by TB program ART.48,49 ART can usually be continued in people
staff during normal business hours.42 Studies eval- experiencing IRIS. Corticosteroids may be benefi-
uating video DOT, either real-time or recorded, cial in people with central nervous system disease
have found that adherence is comparable to in- or severe IRIS.50
person DOT with high levels of patient
satisfaction.42 Renal Disease
Patients are usually seen monthly at in-clinic
Patients with end-stage renal disease are at
visits to monitor their response to treatment and
increased risk for TB and for poor outcomes
assess for drug-related toxicity. Visual acuity and
from disease.51 PZA and EMB are cleared by the
color vision should be checked for patients taking
kidneys and the doses must be adjusted when
EMB. Sputum cultures should be collected each
creatinine clearance is less than 30 mL/min. The
month until there are 2 consecutive specimens
recommendation is to give the usual daily dose 3
that are negative. NAATs are not recommended
times per week after hemodialysis.18 INH and RIF
at this time for monitoring response to treatment
are metabolized by the liver so do not need to be
or suspected failure or relapse. These tests may
adjusted. For patients treated with a fluoroquino-
detect TB DNA for months or even years after
lone, levofloxacin is cleared renally and can be
effective treatment in patients who are otherwise
dosed intermittently when the creatinine clearance
clinically well and have negative cultures.43
is less than 50 mL/min. Moxifloxacin is cleared by
Laboratory monitoring for toxicity is not routinely
the liver so can be given without dose adjustment.
needed. Liver function tests should be checked at
least monthly in patients with known liver disease
Liver Disease
or baseline abnormalities. They should also be
checked immediately in anyone who develops Patients with liver disease are also at risk for com-
nausea, vomiting, abdominal pain, loss of appe- plications due to TB and TB treatment. INH, RIF,
tite, or jaundice. Other laboratory tests should be and PZA can all cause drug-induced liver injury.
done as needed based on symptoms or other Reports suggest that PZA and INH are associated
medical conditions. with more liver toxicity than RIF. RIF should be
suspected when the total bilirubin and alkaline
phosphatase are elevated out of proportion to
SPECIAL CIRCUMSTANCES
the transaminases. Nevertheless, RIF is commonly
Tuberculosis and HIV
tried without the other drugs after the liver function
The approach to treating TB in people living with tests improve regardless of the pattern because it
HIV is similar to treatment of HIV-negative individ- is the most effective drug for curing TB with the
uals.44 Relapse rates after 6 months of treatment shortest treatment duration.
of drug-susceptible pulmonary TB in people Optimal dosing of INH and RIF in patients with
receiving treatment of HIV are low and comparable chronic liver disease is unknown and the choice
to people without HIV.45 In individuals who are not of treatment depends on the degree of functional
started on antiretroviral therapy (ART) and are impairment. Patients with chronic liver disease
highly immunocompromised, extending treatment without cirrhosis and whose baseline transami-
to 9 months may be beneficial.18,46 The most nases are less than 3 times the upper limit of
important consideration is often the drug-drug in- normal can often be treated with standard therapy
teractions with RIF. Rifabutin may be needed to and close monitoring. PZA may be avoided in pa-
replace RIF with many ART combinations. tients with moderate liver disease or higher risk of
 Management of Pulmonary Tuberculosis 33

hepatotoxicity. Moxifloxacin is cleared by the liver by identifying and evaluating those at risk but
and has rarely been associated with hepatotoxicity with minimal or no symptoms is effective for diag-
so levofloxacin may be preferred. nosing TB earlier but requires more resources.
In patients with advanced liver disease, The biggest advance in diagnostic testing has
choosing a safe regimen can be difficult. Levoflox- been with the NAATs that are more accurate
acin, EMB, and injectable medications are gener- than AFB smears for diagnosing pulmonary TB.
ally safe. RIF is often included with close clinical The GeneXpert Omni is battery operated so can
and laboratory monitoring given its importance be used outside of a laboratory. This moves the
for curing TB. Cycloserine, linezolid, and clofazi- technology closer to being a point-of-care test
mine are other TB treatments that have a low risk for rapid TB diagnosis. Continued improvements
for hepatotoxicity but should be considered care- that minimize the need for specimen processing
fully given their potential nonhepatic toxicity. and decrease the cost are still needed.
Drug-susceptible pulmonary TB is curable but re-
Adjunctive Steroids and Tumor Necrosis Factor quires a combination of medications and a lengthy
a Inhibitors treatment course that has not changed in approxi-
Active TB can cause severe inflammation at the mately 40 years. Increasing prevalence and severity
site of disease that may worsen after starting TB of resistant TB threaten to reverse the recent de-
treatment. This acute worsening has been termed clines in the global TB burden. DOT remains the
a paradoxical reaction and is clinically similar to standard of care for treating pulmonary TB to mini-
IRIS in people with HIV after starting ART. Depend- mize the risks of acquired drug resistance and
ing on the site and extent of TB disease, these re- ongoing community transmission. Digital technolo-
actions can be severe and even life threatening. gies offer a more patient-centered alternative to in-
Systemic steroids have been used successfully person DOT for monitoring adherence. New and
to treat and prevent the complications of TB repurposed drugs are providing minimal improve-
IRIS, as discussed previously.50 In patients with ment in treating MDR TB and XDR TB. Outcomes
TB meningitis, steroids have been shown to are still unacceptably poor and treatment remains
decrease the risk of death and are recommended difficult and costly for patients and health care pro-
as adjunctive therapy tapered over 6 weeks to viders. Progress toward reducing the global burden
8 weeks.52,53 Steroid therapy was commonly of TB will continue to be slow without better diag-
used in patients with TB pericarditis but a recent nostic tools and medications that are well tolerated,
randomized controlled trial showed no benefit safe, and effective, with a shorter duration than cur-
and it is no longer recommended by the US guide- rent treatment.
lines.18,54 Adjunctive steroids have been used
sporadically to treat severe pulmonary or miliary REFERENCES
disease but its exact role in these forms of disease
has not been established. Case reports have 1. World Health Organization. Global tuberculosis
described using tumor necrosis factor a inhibitors report. Geneva (Switzerland): WHO; 2017.
to treat severe TB IRIS that was unresponsive to 2. Zink AR, Sola C, Reischl U, et al. Characterization of
steroids and warrants further study.55,56 Mycobacterium tuberculosis complex DNAs from
Egyptian mummies by spoligotyping. J Clin Micro-
SUMMARY biol 2003;41:359–67.
3. Houben RM, Dodd PJ. The global burden of latent
Pulmonary TB remains a common disease globally tuberculosis infection: a re-estimation using mathe-
and is usually a subacute illness. The primary strat- matical modelling. PLoS Med 2016;13:e1002152.
egy for diagnosing TB in high-burden, low- 4. Stewart RJ, Tsang CA, Pratt RH, et al. Tuberculosis
resource settings is through passive case finding — United States, 2017. MMWR Morb Mortal Wkly
that involves evaluating people who present for Rep 2018;67:317–23.
care due to symptoms. Lack of easy access to 5. Tiemersma EW, van der Werf MJ, Borgdorff MW,
health care, stigma associated with TB, and fear et al. Natural history of tuberculosis: duration and fa-
of not being able to work often prevent people tality of untreated pulmonary tuberculosis in HIV
from seeking care early. Delays lead to progres- negative patients: a systematic review. PLoS One
sive disease, permanent lung damage, and lifelong 2011;6:e17601.
disability or death. Delays also increase the risk of 6. Ravimohan S, Kornfeld H, Weissman D, et al. Tuber-
transmission to others. Addressing the system- culosis and lung damage: from epidemiology to path-
level, social, and financial barriers to seeking ophysiology. Eur Respir Rev 2018;27 [pii:170077].
care are important to decreasing morbidity and 7. Harries AD, Satyanarayana S, Kumar AM, et al.
mortality from pulmonary TB. Active case finding Epidemiology and interaction of diabetes mellitus
34 Belknap

and tuberculosis and challenges for care: a review. 20. Tuberculosis Trials Consortium. Once-weekly rifa-
Public Health Action 2013;3:S3–9. pentine and isoniazid versus twice-weekly rifampin
8. Daley C, Gotway M, Jasmer R. Radiographic and isoniazid in the continuation phase of therapy
manifestations of tuberculosis. 2011. Available at: for drug-susceptible pulmonary tuberculosis: a pro-
http://www.currytbcenter.ucsf.edu/topics-interest/tb- spective, randomized clinical trial among HIV-
radiology. Accessed October 7, 2018. negative persons. Lancet 2002;360:528–34.
9. Coleman MT, Chen RY, Lee M, et al. PET/CT imaging 21. Khan A, Sterling TR, Reves R, et al. Lack of weight
reveals a therapeutic response to oxazolidinones in gain and relapse risk in a large tuberculosis treatment
macaques and humans with tuberculosis. Sci Transl trial. Am J Respir Crit Care Med 2006;174:344–8.
Med 2014;6:265ra167. 22. Gegia M, Winters N, Benedetti A, et al. Treatment of
10. Lin PL, Maiello P, Gideon HP, et al. PET CT identifies isoniazid-resistant tuberculosis with first-line drugs:
reactivation risk in cynomolgus macaques with a systematic review and meta-analysis. Lancet
latent M. tuberculosis. PLoS Pathog 2016;12: Infect Dis 2017;17:223–34.
e1005739. 23. Schechter MC, Bizune D, Kagei M, et al. Time to
11. Vorster M, Sathekge MM, Bomanji J. Advances in sputum culture conversion and treatment outcomes
imaging of tuberculosis: the role of (1)(8)F-FDG among patients with isoniazid-resistant tuberculosis
PET and PET/CT. Curr Opin Pulm Med 2014;20: in Atlanta, Georgia. Clin Infect Dis 2017;65:1862–71.
287–93. 24. Jindani A, Harrison TS, Nunn AJ, et al. High-dose ri-
12. Lewinsohn DM, Leonard MK, LoBue PA, et al. Offi- fapentine with moxifloxacin for pulmonary tubercu-
cial American Thoracic Society/Infectious Diseases losis. N Engl J Med 2014;371:1599–608.
Society of America/Centers for Disease Control 25. World Health Organization. WHO treatment guide-
and Prevention clinical practice guidelines: diag- lines for drug resistant tuberculosis. Geneva
nosis of tuberculosis in adults and children. Clin (Switzerland): WHO; 2016.
Infect Dis 2017;64:e1–33. 26. Drug-Resistant Tuberculosis, a survival guide for
13. Mase SR, Ramsay A, Ng V, et al. Yield of serial clinicians 3rd edition. 2016. Available at: http://www.
sputum specimen examinations in the diagnosis of currytbcenter.ucsf.edu/products/view/drug-resistant-
pulmonary tuberculosis: a systematic review. Int J tuberculosis-survival-guide-clinicians-3rd-edition.
Tuberc Lung Dis 2007;11:485–95. Accessed October 7, 2018.
14. Brown M, Varia H, Bassett P, et al. Prospective study 27. Kang YA, Shim TS, Koh WJ, et al. Choice between
of sputum induction, gastric washing, and bron- levofloxacin and moxifloxacin and multidrug-
choalveolar lavage for the diagnosis of pulmonary resistant tuberculosis treatment outcomes. Ann Am
tuberculosis in patients who are unable to expecto- Thorac Soc 2016;13:364–70.
rate. Clin Infect Dis 2007;44:1415–20. 28. Dawson R, Diacon AH, Everitt D, et al. Efficiency and
15. Luetkemeyer AF, Firnhaber C, Kendall MA, et al, safety of the combination of moxifloxacin, pretoma-
AIDS Clinical Trials Group A5295 and Tuberculosis nid (PA-824), and pyrazinamide during the first 8
Trials Consortium Study 34 Teams. Evaluation of weeks of antituberculosis treatment: a phase 2b,
Xpert MTB/RIF versus AFB smear and culture to open-label, partly randomised trial in patients with
identify pulmonary tuberculosis in patients with sus- drug-susceptible or drug-resistant pulmonary tuber-
pected tuberculosis from low and higher prevalence culosis. Lancet 2015;385:1738–47.
settings. Clin Infect Dis 2016;62:1081–8. 29. Reuter A, Tisile P, von Delft D, et al. The devil we know: is
16. Gui X, Xiao H. Diagnosis of tuberculosis pleurisy the use of injectable agents for the treatment of MDR-
with adenosine deaminase (ADA): a systematic re- TB justified? Int J Tuberc Lung Dis 2017;21:1114–26.
view and meta-analysis. Int J Clin Exp Med 2014;7: 30. Chang KC, Nuermberger E, Sotgiu G, et al. New
3126–35. drugs and regimens for tuberculosis. Respirology
17. Diel R, Loddenkemper R, Nienhaus A. Evidence- 2018. https://doi.org/10.1111/resp.13345.
based comparison of commercial interferon- 31. Van Deun A, Maug AK, Salim MA, et al. Short, highly
gamma release assays for detecting active TB: a effective, and inexpensive standardized treatment of
metaanalysis. Chest 2010;137:952–68. multidrug-resistant tuberculosis. Am J Respir Crit
18. Nahid P, Dorman SE, Alipanah N, et al. Official Amer- Care Med 2010;182:684–92.
ican Thoracic Society/Centers for Disease Control 32. Aung KJ, Van Deun A, Declercq E, et al. Successful
and Prevention/Infectious Diseases Society of Amer- ’9-month Bangladesh regimen’ for multidrug-
ica clinical practice guidelines: treatment of drug- resistant tuberculosis among over 500 consecutive
susceptible tuberculosis. Clin Infect Dis 2016;63: patients. Int J Tuberc Lung Dis 2014;18:1180–7.
e147–95. 33. Sotgiu G, Tiberi S, Centis R, et al. Applicability of the
19. World Health Organization. Guidelines for treatment shorter ’Bangladesh regimen’ in high multidrug-
of drug-susceptible tuberculosis and patient care, resistant tuberculosis settings. Int J Infect Dis
2017 update. Geneva (Switzerland): WHO; 2017. 2017;56:190–3.
 Management of Pulmonary Tuberculosis 35

34. Berry C, Yates TA, Seddon JA, et al. Efficacy, safety antiretroviral therapy for HIV-positive adults with
and tolerability of linezolid for the treatment of XDR- newly diagnosed pulmonary tuberculosis (TB-
TB: a study in China. Eur Respir J 2016;47:1591–2. HAART): a prospective, international, randomised,
35. Agyeman AA, Ofori-Asenso R. Efficacy and safety placebo-controlled trial. Lancet Infect Dis 2014;14:
profile of linezolid in the treatment of multidrug- 563–71.
resistant (MDR) and extensively drug-resistant 46. Nahid P, Gonzalez LC, Rudoy I, et al. Treatment out-
(XDR) tuberculosis: a systematic review and meta- comes of patients with HIV and tuberculosis. Am J
analysis. Ann Clin Microbiol Antimicrob 2016;15:41. Respir Crit Care Med 2007;175:1199–206.
36. Borisov SE, Dheda K, Enwerem M, et al. Effective- 47. Torok ME, Yen NT, Chau TT, et al. Timing of initiation
ness and safety of bedaquiline-containing regimens of antiretroviral therapy in human immunodeficiency
in the treatment of MDR- and XDR-TB: a multicentre virus (HIV)–associated tuberculous meningitis. Clin
study. Eur Respir J 2017;49 [pii:1700387]. Infect Dis 2011;52:1374–83.
37. Burman W, Benator D, Vernon A, et al. Acquired rifa- 48. Naidoo K, Yende-Zuma N, Padayatchi N, et al. The
mycin resistance with twice-weekly treatment of HIV- immune reconstitution inflammatory syndrome after
related tuberculosis. Am J Respir Crit Care Med antiretroviral therapy initiation in patients with tuber-
2006;173:350–6. culosis: findings from the SAPiT trial. Ann Intern Med
38. Johnston JC, Campbell JR, Menzies D. Effect of 2012;157:313–24.
intermittency on treatment outcomes in pulmonary 49. Lawn SD, Myer L, Bekker LG, et al. Tuberculosis-
tuberculosis: an updated systematic review and associated immune reconstitution disease: inci-
metaanalysis. Clin Infect Dis 2017;64:1211–20. dence, risk factors and impact in an antiretroviral
39. Gopalan N, Santhanakrishnan RK, Palaniappan AN, treatment service in South Africa. AIDS 2007;21:
et al. Daily vs intermittent antituberculosis therapy 335–41.
for pulmonary tuberculosis in patients with hiv: a ran- 50. Meintjes G, Wilkinson RJ, Morroni C, et al. Random-
domized clinical trial. JAMA Intern Med 2018;178: ized placebo-controlled trial of prednisone for
485–93. paradoxical tuberculosis-associated immune recon-
40. DeMaio J, Schwartz L, Cooley P, et al. The applica- stitution inflammatory syndrome. AIDS 2010;24:
tion of telemedicine technology to a directly 2381–90.
observed therapy program for tuberculosis: a pilot 51. Baghaei P, Marjani M, Tabarsi P, et al. Impact of
project. Clin Infect Dis 2001;33:2082–4. chronic renal failure on anti-tuberculosis treatment
41. Krueger K, Ruby D, Cooley P, et al. Videophone uti- outcomes. Int J Tuberc Lung Dis 2014;18:352–6.
lization as an alternative to directly observed ther- 52. Critchley JA, Young F, Orton L, et al. Corticosteroids
apy for tuberculosis. Int J Tuberc Lung Dis 2010; for prevention of mortality in people with tubercu-
14:779–81. losis: a systematic review and meta-analysis. Lancet
42. Garfein RS, Collins K, Munoz F, et al. Feasibility of Infect Dis 2013;13:223–37.
tuberculosis treatment monitoring by video directly 53. Prasad K, Singh MB, Ryan H. Corticosteroids for
observed therapy: a binational pilot study. Int J Tu- managing tuberculous meningitis. Cochrane Data-
berc Lung Dis 2015;19:1057–64. base Syst Rev 2016;(4):CD002244.
43. Theron G, Venter R, Smith L, et al. False-positive 54. Mayosi BM, Ntsekhe M, Bosch J, et al. Prednisolone
Xpert MTB/RIF results in retested patients with previ- and Mycobacterium indicus pranii in tuberculous
ous tuberculosis: frequency, profile, and prospective pericarditis. N Engl J Med 2014;371:1121–30.
clinical outcomes. J Clin Microbiol 2018;56(3). 55. Wallis RS, van Vuuren C, Potgieter S. Adalimumab
e01696–17. treatment of life-threatening tuberculosis. Clin Infect
44. Haas MK, Daley CL. Mycobacterial lung disease Dis 2009;48:1429–32.
complicating HIV infection. Semin Respir Crit Care 56. Hsu DC, Faldetta KF, Pei L, et al. A paradoxical treat-
Med 2016;37:230–42. ment for a paradoxical condition: infliximab use in
45. Mfinanga SG, Kirenga BJ, Chanda DM, et al. Early three cases of mycobacterial IRIS. Clin Infect Dis
versus delayed initiation of highly active 2016;62:258–61.