310

Management of Multidrug-Resistant Tuberculosis

Charles L. Daley1,2,3 Jose A. Caminero4,5

1 Division of Mycobacterial and Respiratory Infections, National Jewish Address for correspondence Charles L. Daley, MD, Division of
Health, Denver, Colorado Mycobacterial and Respiratory Infections, National Jewish Health,
2 Division of Pulmonary Sciences and Critical Care, Department of 1400 Jackson Street, Denver, CO 80206 (e-mail: daleyc@njhealth.org).
Medicine, University of Colorado Denver, Aurora, Colorado
3 Division of Infectious Diseases, Department of Medicine,
University of Colorado Denver, Aurora, Colorado
4 Servicio de Neumologia, Hospital General de Gran Canaria,
Las Palmas, Canary Islands, Spain
5 International Union against Tuberculosis and Lung Disease
(The Union), Paris, France

Semin Respir Crit Care Med 2018;39:310–324.

pas
¡

}
Abstract Drug-resistant strains of Mycobacterium tuberculosis pose a major threat to global
tuberculosis control. Despite the availability of curative antituberculosis therapy for

Downloaded by: Serials Unit - Periodicals. Copyrighted material.

nuevos casos
4.1%
TB MDR
-

•
\ nearly half a century, inappropriate and inadequate treatment of tuberculosis, as well
retratados 19%
as unchecked transmission of M. tuberculosis, has resulted in alarming levels of drug-
ti: * "
resistant tuberculosis. The World Health Organization (WHO) estimates that there
were 600,000 cases of multidrug-resistant tuberculosis (MDR-TB)/rifampin-resistant
(RR) tuberculosis in 2016, defined as strains that are resistant to at least isoniazid and
Keywords rifampicin. Globally, WHO estimates that 4.1% of new tuberculosis cases and 19% of
► tuberculosis retreatment cases have MDR-TB. By the end of 2016, 123 countries had reported at
► drug-resistant least one case of extensively drug-resistant strains, which are MDR-TB strains that have
tuberculosis acquired additional resistance to fluoroquinolones and at least one second-line
► MDR-TB injectable. It is estimated that only 22% of all MDR-TB cases are currently receiving
► XDR-TB therapy. This article reviews the management of MDR/RR-TB and updates recommen-
► management dations regarding the use of shorter course regimens and new drugs.

Drug-resistant strains of Mycobacterium tuberculosis pose a

major threat to global tuberculosis (TB) control. Despite the
availability of curative anti-TB therapy, inappropriate and
µ ÷:
TB strains that have acquired additional resistance to fluor-
oquinolones and at least one second-line injectable. WHO
estimates that 6.2% of MDR-TB cases have XDR-TB.
:c
XDR
inadequate treatment, as well as unchecked transmission, Management of drug-resistant TB begins with the identifi- -

has allowed drug-resistant strains of M. tuberculosis to reach
alarming levels. The World Health Organization (WHO) esti-
mates that there were 600,000 cases of multidrug-resistant TB
(MDR-TB)/rifampin-resistant TB (RR-TB) in 2016, with MDR-
TB, defined as strains that are resistant to at least isoniazid
(INH) and rifampicin (RIF), accounting for 82% of the total.1 The
WHO estimates that 4.1% of new TB cases and 19% of retreat-
cation of cases. However, accurate diagnosis of drug-resistance
remains a barrier to control. The END TB Strategy calls for
-
universal access to drug susceptibility testing (DST) including
DST for at least RIF for all TB cases, plus DST for at least
fluoroquinolones and second-line injectable agents among
cases with RIF resistance.2 However, in 2016, only 41% of
laboratory-confirmed TB patients notified globally were tested dx

÷:*
ment cases have MDR-TB globally. The countries with the for RIF resistance, up from 11% in 2012. The development of
largest numbers of MDR/RR-TB, accounting for 47% of the new rapid molecular tests such as the Xpert MTB/RIF assay3
and line probe assays4 allow for the diagnosis of INH and RIF

!
global total, are China, India, and the Russian Federation. By the
end of 2016, 123 countries had reported at least one case of
extensively drug-resistant strains (XDR-TB), which are MDR-
resistance within a matter of hours, and these assays have
-
accounted for some of the recent increases in testing.

:*
Issue Theme Mycobacterial Diseases:
Evolving Concepts; Guest Editors: Patrick
A. Flume, MD, and Kevin L. Winthrop,
MD, MPH
Copyright © 2018 by Thieme Medical DOI https://doi.org/
Publishers, Inc., 333 Seventh Avenue, 10.1055/s-0038-1661383.
New York, NY 10001, USA. ISSN 1069-3424.
Tel: +1(212) 584-4662.
 Management of Multidrug-Resistant Tuberculosis Daley, Caminero 311

Among the 490,000 incident cases of MDR-TB in 2016, only
153,000 (31%) were detected, and of these, 130,000 (85%) were
started on MDR-TB treatment.1 It is estimated that only 22% of
the 600,000 MDR/RR-TB cases started second-line treatment.
59%0 ← Treatment success is estimated to be approximately 54%, with
16% of patients dying, 8% failing, and 21% lost to follow-up or not
evaluated. Treatment success in XDR-TB patients is only 30%.
Group A Drugs
Group A drugs include the later-generation fluoroquinolones
(high-dose levofloxacin, moxifloxacin, and gatifloxacin), which
are among the most important SLDs available to treat MDR-TB.
Fluoroquinolones have bactericidal and sterilizing activity and
are very well tolerated, and resistance to these drugs is asso-
ciated with higher rates of treatment failure and death.5,12

3_
↳
Inclusion of fluoroquinolones in the treatment regimen is

Treatment of Multidrug-Resistant
Tuberculosis
Treatment of MDR-TB is challenging because it requires the
D-
administration of at least five anti-TB drugs, many associated
with significant adverse reactions, for at least 9 to 20 months.
III:
associated with improved outcomes.13,14 Resistance varies
depending on which fluoroquinolone and concentration are
used to define resistance. When ofloxacin (2 μg/mL) was used in
a WHO study, resistance ranged from 12.3% in Gauteng, South
Africa, to 30.7% in Minsk, Belarus.15 Levofloxacin has been
shown to be more effective than ofloxacin in both ofloxacin- -

susceptible and ofloxacin-resistant strains of M. tuberculosis.16

÷:* ±
Selection of drugs to make up the regimen should be based
. -
-

on the past treatment history, known patterns of resistance, Furthermore, early bactericidal studies have demonstrated that

yandex.IN
_ -

÷÷÷÷¥P
and DST data, when available. At least two drugs should have levofloxacin, when given at high doses (1,000 mg/day), is asso-
-

÷÷¥I÷
good bactericidal activity, and one to two drugs should have ciated with early bactericidal activity (EBA) activity equivalent to
-
""

good sterilizing activity, to assure cure and avoid relapse.5,6 moxifloxacin.17,18 Moxifloxacin appears to have better steriliz-
The WHO updated its guidelines for the programmatic ing capacity than ofloxacin in animal models.19 In a retro-

Downloaded by: Serials Unit - Periodicals. Copyrighted material.

management of drug-resistant TB in 2016,7 with additional
updates related to new drugs including bedaquiline8 and
delamanid,9,10 as well as shorter course regimens.11
WHO Regrouping of Second-Line Tuberculosis Drugs
The WHO regrouped the second-line drugs (SLDs) based on
current evidence on their effectiveness and safety (►Table 1).7
In the new scheme, drugs are divided into groups A to D, with
drugs listed in order of general preference in groups A to C.
Group A includes the later-generation fluoroquinolones, group
B includes the injectables, and group C includes other core drugs
such as ethionamide/prothionamide, cycloserine/terizidone,
spective study from South Korea that included 171 patients,
treatment success was similar between levofloxacin and moxi-
floxacin, although only 48 patients received the latter drug
(normal dose), and the patients who received moxifloxacin
had isolates with resistance to significantly more drugs and
higher incidence of ofloxacin-resistant than those who took
levofloxacin (78.9 vs. 83.3%; p ¼ 0.42).20 The WHO recommends
a-
that a later-generation fluoroquinolone (levofloxacin, moxiflox-
acin, gatifloxacin) should be used in all MDR-TB regimens.7 ⑧
Up to half of strains resistant to ofloxacin retain some
=
susceptibility to moxifloxacin21,22 and to high-dose levo-
floxacin.22 Zignol et al15 reported that among 282 strains of
consideradoresistant to ofloxacin at 2 μg/mL, 72%
-

linezolid, and clofazimine. Group D includes “add-on” agents

-
that are divided into three subgroups: D1, pyrazinamide (PZA),
ethambutol (EMB), and high-dose INH; D2, bedaquiline and
delamanid; D3, para-aminosalicylic acid (PAS), imipenem-
cilastatin, meropenem, amoxicillin-clavulanate, and thioaceta-
zone (if HIV-negative). Appropriate dosing of these drugs is
displayed in ►Table 2.
M. tuberculosis deemed
were resistant to moxifloxacin at 0.5 μg/mL, but only 7% were
resistant to moxifloxacin and 2% to gatifloxacin at a concen-
tration of 2 μg/mL. Because some strains of M. tuberculosis
a
have demonstrated activity despite resistance to ofloxacin,
and this is the most commonly used fluoroquinolone to
define phenotypic resistance, it is currently recommended

Table 1 Antimycobacterial drugs recommended for the treatment of MDR/RR-TB

Group A Group B Group C Group D

Fluoroquinolone Second-line injectables Other core second-line Add-on agents
drugs
Levofloxacin Amikacin Ethionamide/prothionamide D1:
Moxifloxacin Capreomycin Cycloserine/Terizidone Pyrazinamide
Gatifloxacin Kanamycin Clofaziminea Ethambutol
(streptomycin) Linezolida High-dose INH
D2:
Bedaquiline
Delamanid
D3:
P-aminosalicylic acida
Imipenem/meropenemb
Amoxicillin/clavulanate
(thioacetazone)
a
Clofazimine and linezolid were moved from the previous group 5 designation to group C. P-aminosalicylic acid was moved from the previous group 4
designation to group D3. bCarbapenems should be combined with clavulanate, which is available as amoxicillin/clavulanate.

Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018

 312 Management of Multidrug-Resistant Tuberculosis Daley, Caminero

Table 2 Drugs and dosages

Group/drugs Daily dose Comments related to the treatment of MDR-TB

Group A: fluoroquinolones
Levofloxacin 15 mg/kg Equal EBA to moxifloxacin, better PD environment
Moxifloxacin 7.5–10 mg/kg Best combination of EBA and sterilizing capacity
Group B: injectable antituberculosis drugs
Streptomycin 15 mg/kg Not recommended for MDR-TB
Kanamycin 15 mg/kg First choice given cost and availability
Amikacin 15 mg/kg Used when kanamycin is not available
Capreomycin 15 mg/kg Typically reserved for the retreatment of MDR-TB or XDT-TB
Group C: other core drugs
Ethionamide/ 15 mg/kg Better activity and tolerance than cycloserine
prothionamide
Cycloserine/terizidone 15 mg/kg Active agent but with significant adverse reactions
Linezolid 600 mg Good activity, frequent toxicity at higher doses
Clofazimine 100 mg Good activity, not readily available although part

Downloaded by: Serials Unit - Periodicals. Copyrighted material.

of shorter course regimen
Group D: add-on agents
Ethambutol 15–25 mg/kg Should be considered to strengthen regimen
Pyrazinamide 30 mg/kg Should be included in all MDR-TB regimens
High-dose isoniazid 10–15 mg/kg Should be considered to strengthen regimen and in
shorter course regimen
Bedaquiline 400 mg daily for 2 weeks Excellent activity and well tolerated
then 200 mg three
times weekly
Delamanid 200 mg twice daily Excellent activity and well tolerated
P-aminosalicylic 150 mg/kg Little activity and poorly tolerated
acid (acid salt)
Amoxicillin with 875/125 mg every 12 hour Weak activity, readily available, well tolerated;
clavulanate combined with carbapenems
Imipenem/cilastatin, 500–1,000 mg every 6 hour Good activity, intravenous only
meropenem
Thioacetazone 150 mg Not readily available, frequent toxicity in HIVþ

Abbreviations: EBA, early bactericidal activity; MDR-TB, multidrug-resistant tuberculosis; PD, pharmacodynamic; XDT-TB, extensively drug-resistant
tuberculosis.

&
to include moxifloxacin in any treatment regimen for XDR- Group B Drugs
TB, especially at high dose, and to test systematically this Group B drugs include aminoglycosides (streptomycin, kana-

:D -
Recomienda
drug at a high concentration. mycin, and amikacin) and polypeptides (capreomycin and

:{
WHO recommends that GenoType MDRTBsl, a rapid mole- viomycin).24 These drugs are bactericidal and act primarily

÷÷÷÷÷
" _

cular test, should be performed on all M. tuberculosis strains extracellularly, although some (very little) intracellular
with RR-TB or MDR-TB.23 This assay has a high sensitivity and activity has been demonstrated.5,25 All of the injectables

iii.
a very high specificity to detect resistance to the fluoroqui- have similar levels of in vitro activity and similar adverse
nolones (mutations in the genes gyrA and gyrB) and SLD reaction profiles; however, capreomycin is more likely to

sentsibeet
-

✓
. injectables (mutations in the genes rrs and eis). However,
there are some specific mutations that do not confer resis-
cause electrolyte disturbances such as hypokalemia and
hypomagnesemia.26–28 Unfortunately, all of these agents
-
✓
espeihiot tance to all fluoroquinolones. For instance, mutation in GyrA are ototoxic and have been reported to produce hearing
a detectar Asp94Gly confers high resistance to all the fluoroquinolones ②
loss in up to 50% of MDR-TB patients who received TODOS
Rt ntnyet .
(the same for all 94 mutations except Asp94Ala). However, them.29,30 Although cross-resistance among the agents is Animo
Mutaciones mutation in GyrA Ala90Val can confer low or high resistance common, our understanding of the cross-resistance between RAM :

rototoxiuidad
^ A to fluoroquinolones, and, for this reason, it is better to add the injectables has evolved with improved understanding of
iii. rriiis high-dose moxifloxacin or gatifloxacin.
- the mutational causes of resistance. 24,31
Based on recent
FIÉ .
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
 *÷
:
Management of Multidrug-Resistant Tuberculosis
studies,31–34 it appears that isolates acquiring resistance to
streptomycin usually remain susceptible to kanamycin, ami-
kacin, and capreomycin. Isolates acquiring resistance to
capreomycin are usually susceptible to kanamycin and ami-
kacin. Most isolates that acquire resistance to amikacin are
-
also resistant to both kanamycin and capreomycin, and
isolates acquiring resistance to kanamycin show different
levels of cross-resistance to amikacin and capreomycin. The
Üw¡É
WHO recommends that all adult patients with MDR/XDR-TB
receive an injectable agent as part of their initial therapy,
whether in a shorter course or conventional regimen.7
!::
As with the fluoroquinolones, GenoType MDRTBsl can
identify specific mutations, some of which do not confer
resistance to all the second-line injectables. For instance,
:÷:
mutations in rrs1401 or 1484 usually confer high-level
resistance to all injectables; however, streptomycin may still
be active if there is no rpSL mutation. On the contrary,
÷:
mutations in rrs1402 can confer resistance to kanamycin
and capreomycin, but amikacin may retain susceptibility.
And mutations only in eis confer low-level resistance to
kanamycin but susceptibility to amikacin.
Group C Drugs
These core SLDs include thioamides (ethionamide and
prothionamide), cycloserine (and its derivative terizidone),
clofazimine, and linezolid.7 Based on efficacy, toxicity, and
cost, the following order of introduction is recommended by
the authors: linezolid, thioamide, clofazimine, and cycloser-
ine. The WHO analysis noted that the association with cure
was higher with ethionamide than with cycloserine.7
Linezolid
Linezolid is an oxazolidinone antibiotic with moderate bac-
tericidal and possibly good sterilizing activity.5 It can be used
to strengthen conventional MDR-TB regimens or as one of
the core drugs for -5pre-XDR-TB and XDR-TB treatment. A
-
systematic review and meta-analysis concluded that line-
zolid appeared to provide added benefit to multidrug regi-
mens, although drug-related toxicity was common.35 Twelve
studies from 11 countries with 207 patients were included in
the review, and individual data from 121 patients were
included in the meta-analysis. Culture conversion was
reported to have occurred in 100/107 (93.5%) patients
receiving individualized linezolid-containing MDR-TB regi-
mens. Eighty-two percent of the patients were treated
-
successfully, and there was no difference in outcomes
when stratified by doses above or below 600 mg daily. Based
on this review, the authors concluded that #600 mg/day
(single or divided doses) was the best option currently.
Another systematic review and meta-analysis that included
11 studies with 148 patients reported a pooled treatment
success of 67.9% and again noted no significant difference in
outcomes based on whether the patient received #600 mg/
day or >600 mg/day.36
One of the studies reviewed in each of the preceding
systematic reviews was that by Koh et al,37 who reported
good outcomes and possibly less peripheral neuropathy
when using 300 mg/day in a group of highly resistant MDR
Daley, Caminero 313
patients who had been on treatment for median of 40 months
prior to initiating linezolid. A study from Shanghai, China,
which was not included in either review, reported that
culture conversion occurred in all 18 MDR-TB (including 15
XDR-TB) patients who received linezolid (900–1,200 mg
intravenously) as part of their drug regimen.38 At the end
of the study, 9 of the 18 patients had achieved treatment
success.
A randomized phase II clinical trial in patients with XDR-
TB demonstrated high rates of culture conversion.39,40
Subsequently, Tang et al41 published a prospective, multi-
center, randomized study to evaluate the efficacy, safety,
and tolerability of linezolid in XDR-TB: 65 patients were
randomly assigned to a 2-year individually based regimen
with or without linezolid. The linezolid therapy group was
given linezolid at a start dose of 1,200 mg/day for a period
of 4 to 6 weeks, and this was then followed by a dose of 300
to 600 mg/day. The proportion of sputum culture conver-
sions in the linezolid group was 78.8% by 24 months,
significantly higher than that in the control group (37.6%;
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
p ¼ 0.001), and the treatment success rate in the linezolid
group was 69.7%, significantly higher than that in the
control group (34.4%; p ¼ 0.004). However, 27 (81.8%)
patients had clinically significant adverse events (AEs) in
the linezolid group, of whom 25 (93%) patients had events
that were possibly or probably related to linezolid. Most AEs
resolved after reducing the dosage or temporarily disconti-
nuing linezolid.
Unfortunately, toxicity has been common in almost all
series to date. In the systematic review and meta-analysis
published by Migliori et al,42 the authors reported that AEs
occurred in almost 60% of patients receiving linezolid, of
KEII
which 68% were considered major AEs including anemia
(38%), peripheral neuropathy (46%), gastrointestinal disor-
ders (17%), optic neuritis (13%), and thrombocytopenia
(12%). The frequency of AEs was significantly higher when
the dose of linezolid was -
_
greater than 600 mg daily. Simi-
larly, in the review by Cox and Ford,36 adverse reactions were
common, resulting in discontinuation of linezolid in
approximately 36% of patients. The proportion of AEs neces-
sitating treatment discontinuation varied by dose: 29.5% for
#600 mg and 60.75% for >600 mg (p ¼ 0.05). In the study by
Xu et al,38 17/18 patients suffered side effects, with drug
discontinuation in one patient and dosage adjustments in 17
patients. One approach to managing AEs caused by linezolid
is to start with 600 mg/day and to reduce to 300 mg/day
when the smear is negative or the AEs appear.
Thioamides
The thioamides inhibit mycolic acid biosynthesis, and the
molecular target of ethionamide/prothionamide is inhA.43
Another mutation known to result in resistance is a mutation
in the ethA gene. The thioamides are more bactericidal than
cycloserine or PAS and are less expensive. However, the
bactericidal activity is only moderate, and drug-related toxicity
is often a limiting factor.5• Gastrointestinal disturbance is the
trans
primary adverse effect associated with thioamides, although
hypothyroidism and hepatitis are also possible. Prothionamide
-
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
• Protionamide mejor tolerancia GI q
'
tionamidq
 314 Management of Multidrug-Resistant Tuberculosis
may have better gastrointestinal tolerance that ethionamide,
and thus may be a better choice among the thioamides.44
Clofazimine
Clofazimine is a riminophenazine antibiotic used as a critical
component of multidrug therapy for leprosy. The drug is highly
lipophilic, has good activity in vivo against M. tuberculosis, and
has been shown to be highly active in both acute and chronic
mouse models with less activity demonstrated in guinea pigs
and monkeys.45 A recent study reported that clofazimine had
dose-dependent, sustained bactericidal activity against M.
tuberculosis persisters in a mouse model of chronic TB.46
Clofazimine has been used a part of multidrug regimens for
the treatment of MDR and XDR-TB, but it is not possible to
isolate the activity of clofazimine in such reports.
A recent systematic review and meta-analysis reported
the results of the use of clofazimine in the treatment of drug-
resistant TB.47 The review included 12 studies including
3,489 patients. Treatment success ranged from 16.5% in a
small retrospective study from India to 87.8% in a cohort
study in Niger; the pooled estimate of treatment success was
RAI ( 62%. The most common side effects were
_
disturbances and skin pigmentation.
- gastrointestinal
Two retrospective observational studies compared treat-
ment outcomes of clofazimine- versus nonclofazimine-con-
taining regimens. The first report from Shanghai, China,
reported that no significant difference was seen in treatment
outcomes among the 44 patients who received clofazimine,
although they were more likely to have received a fluoroqui-
nolone and an injectable previously.48 None of the patients
who received clofazimine experienced major side effects
requiring discontinuation; however, the dose was reduced in
20 patients. Adverse reactions occurred in 88% of the patients,
which included darkening of the skin (39), gastrointestinal
adverse reactions (21), ichthyosis (12), and dizziness (1). The
largest series using clofazimine in the treatment of MDR-TB
was published recently from Brazil.49 In this retrospective
study, treatment success was 60.9% and similar to those who
received PZA (64.6%), and adverse reactions were relatively
uncommon. Van Deun et al50,51 reported excellent treatment
outcomes in patients with MDR-TB who received a seven-drug
regimen that included clofazimine. No adverse reactions were
attributed to clofazimine in that study.
A single randomized controlled study assigned 105
patients with MDR-TB in China to receive either an indivi-
dualized regimen with or without clofazimine.41 Use of
clofazimine was associated with cavity closure, accelerated
sputum culture conversion, and higher treatment success
(73.6 vs. 53.8%). While clofazimine appears to have both
in vitro and in vivo activity and is relatively well tolerated,
the drug is currently not widely available as the only source
of quality-assured clofazimine has removed the drug from
the global market. Currently, clofazimine is included on the
mandatory drug list to be used in the shorter MDR/RR-TB
-
regimen recommended by WHO in 20167 and is made
available through the Global Drug Facility. In the United
States, the drug can be acquired through the Food and
Drug Administration.
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
Daley, Caminero
Cycloserine
Cycloserine competitively blocks the enzymes that incorporate
alanine into an alanyl-alanine dipeptide, which is an essential
component in the mycobacterial cell wall.24 Early studies using
cycloserine monotherapy produced rapid clinical response,
and when combined with INH, patients also improved but INH
resistance emerged.52 When combined with ethionamide,
patients usually converted cultures to negative.53 Because of
this documented activity, cycloserine has become an impor-
tant drug for the treatment of MDR/XDR-TB. However, cyclo-
serine has limited bactericidal and sterilizing activity5 and, for
this reason, should be the last choice in group C. Moreover,
-
↳÷:p
cycloserine can be associated with significant psychiatric
adverse reactions such as psychotic reactions and suicidal
ideation. Terizidone, which consists of a double molecule of
cycloserine, may be less toxic is not as available as cycloserine,
and there are few studies describing its effectiveness.24
Group D Drugs
Group D drugs includes three different groups: group D1
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
includes first-line TB drugs such as PZA, EMB, and INH (high-
dose); group D2 includes new drugs such as bedaquiline and
delamanid; and group D3 includes PAS and other drugs pre-
viously included in the previous WHO Group 5, such as the
carbapenems (imipenem or meropenem) plus clavulanic acid
and thioacetazone.24 Which drug to choose should be based on
relative effectiveness, potential for adverse reactions, costs, and
availability.
Group D1
Pyrazinamide
PZA, a prodrug activated by pncA, has potent sterilizing activity5
and is used during the initial phase of anti-TB treatment with
the goal of shortening therapy.24 Resistance to PZA is conferred
by several mutations including mutations in pncA and rpsA.54
Resistance is common among MDR-TB patients ranging from 36
to 81% is some regions.15 Prior to the availability of the
rifamycins and fluoroquinolones, PZA was used in combination
with ethionamide and cycloserine to treat patients with iso-
lates who were resistant to INH and streptomycin.55–57 In
general, treatment outcomes were excellent. In studies describ-
ing the outcome of MDR and XDR-TB patients, inclusion of PZA
in the regimen was associated with better treatment outcomes
than when PZA was not included in the regimen.58,59 In a recent
large meta-analysis supporting the 2011 WHO guidelines,
inclusion of PZA in the treatment regimen was associated
with improved outcomes, although the added benefit was
small.26 In a subsequent analysis, PZA resistance was not
associated with a significant decrease in treatment success
with either the shorter course or conventional regimen.7
Because DST methods for PZA are not as reproducible as those
for INH and RIF, and availability of appropriate methods is often
poor, the WHO recommendations including PZA in all treat-
A
ment regimens throughout the course of MDR-TB therapy,
regardless of DST results, including the shorter MDR/RR-TB
regimens.7 PZA is associated with hepatotoxicity, and therefore
patients should be monitored closely for signs of liver damage.
RAMTLI
 Management of Multidrug-Resistant Tuberculosis
Ethambutol
EMB is one of the least active of the first-line drugs but prevents
the emergence of resistance to companion agents.5 EMB
resistance has been reported to occur in as much as 50 to
60% of MDR-TB cases, but this varies widely.43 Resistance is
conferred by mutations in embB gene, but this only accounts for
60% of resistance. Reproducibility of DST is relatively poor and
not currently recommended by the WHO.60 Current guidelines
recommend consideration of inclusion of EMB in an MDR-TB
regimen, and it is a drug included in the shorter MDR/RR-TB
regimens.7 When used, the drug should not be included as one
of the four effective drugs in the treatment regimen. Higher
-
doses (25 mg/kg) are more active but are also more likely to
£9
produce optic neuritis than standard dosing (15 mg/kg).
L
si alta dosis
Isoniazid
RIM
INH is a prodrug that must be activated by catalase-peroxidase
to be active against M. tuberculosis.43 Mutations within the
zona
katG gene result in moderate to high-level resistance to INH
(minimal inhibitory concentration [MIC] $ 16 μg/mL). Resis-
tant can also occur when mutations occur in inhA and its
promoter region, but this causes lower levels of resistance
(MIC: 2–8 μg/mL).61,62 Mutations in the inhA region also confer
resistance to ethionamide and prothionamide.62–64 Strains
that are moderately to highly resistant to INH due to katG
mutations are likely to be susceptible to ethionamide (prothio-
nomide), whereas strains that demonstrate low-level resis-
tance to INH (due to inhA mutations) are likely to be resistant to
ethionamide and susceptible to high doses of INH.24,65
High-dose INH could potentially overcome low and mod-
erate levels of resistance and provide an active drug for the
treatment of MDR/XDR-TB.66 In a mouse model, an INH dose of
10 mg/kg/day exhibited bactericidal activity against an inhA
promoter mutant, and increasing the dose to 25 mg/kg/day
resulted in bactericidal activity against a katG mutant.67 In a
randomized study evaluating the impact of high-dose INH
(16–18 mg/kg) versus standard dose INH (5 mg/kg), patients
±i÷ -
who received high-dose INH became sputum smear-negative
2.4 times more rapidly and were 2.4 times more likely to
achieve sputum culture conversion at 6 months than those
"
who did not receive high-dose INH.68 INH can produce drug-
induced hepatotoxicity as well as peripheral neuropathy.
←
However, in this study, there was no increase in hepatotoxicity
among patients who received high-dose INH but, there was an
increase in peripheral neuropathy (patients were not given
RAMI supplementation with pyridoxine). Given the low cost of INH
and good reliability of DST, high-dose INH should be consid-
ered as a possible role in treatment of patients with MDR/RR-
TB who have INH resistance documented.24,62,66,69 High-dose
INH is already being used in some treatment programs70 and is
a component of the shorter course MDR-TB regimen.50
Group D2
Bedaquiline
Bedaquiline (TMC-207) is a diarylquinoline that works in a
novel fashion to inhibit mycobacterial adenosine triphosphate
(ATP) synthase.71 Bedaquiline has good bactericidal activity
Daley, Caminero 315
from day 4 onward and good sterilizing activity.5 Resistance to
bedaquiline occurs through mutations in atpE, which encodes
the ATP synthase subunit to which the drug binds.72 Cross-
resistance to clofazimine has been reported through upregula-
tion of MmpL573 and mutations in pepQ.74 In a randomized
placebo-controlled study of 47 patients with MDR-TB, beda-
quiline increased the proportion of patients with conversion
(48 vs. 9%)71 and reduced the time to culture conversion over
24 weeks of observation.75 In addition, none of the patients
who received bedaquiline acquired resistance to ofloxacin
compared with 22% of those on placebo. A subsequent phase
2, open-label, single-arm study reported that 72% of 233 MDR/
XDR-TB patients converted cultures by 120 weeks, and the
bedaquiline-containing regimens were well tolerated.76 A
recent large retrospective study including 428 MDR-TB
patients reported a 71.3% treatment success with interruption
of therapy in only 5.8% of patients.77 Despite these successes,
programmatic uptake of bedaquiline has been slow. Based on a
convenience sample of 36 countries, a total of 10,164 patients
were started on bedaquiline between 2015 and 2017, 80% of
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
whom resided in South Africa (66%) and Russia (14%).78
Delamanid
Delamanid (OPC-67683) is a nitro-dihydro-imidazooxazole
derivative that inhibits mycolic acid synthesis.79 The drug
has good bactericidal and sterilizing activity with activity
against both growing and dormant bacilli.5 In a randomized
placebo-controlled multinational clinical trial, 481 patients
were assigned to receive delamanid 100 mg twice daily,
delamanid 200 mg twice daily, or placebo for 2 months in
combination with an optimized background regimen.79
Culture conversion at 2 months in a liquid culture system
was more likely in patients who received delamanid 100 mg
twice daily (45.4%; p ¼ 0.0008) or delamanid 200 mg twice
daily (41.9%; p ¼ 0.04) than placebo. AEs were distributed
relatively equally across the three groups except that QT eren
prolongation was more common in the delamanid group.
Early results with delamanid in the programmatic setting
have shown high culture conversion rates and low fre-
quency of QTc intervals over 500 millisecond.80–82 Results
from compassionate use in resource-limited settings
demonstrated culture conversion at 6 months in 80% of
patients, with QTc prolongation in only 3.8%.83 As with
bedaquiline, programmatic uptake of delamanid has been
slow. Only 688 patients received delamanid between 2015
and 2017.78
Use of Bedaquiline and Delamanid in Combination
Because of the bactericidal and sterilizing activity of bedaqui-
line and delamanid and their better safety profile, these drugs
-
could replace the second-line injectable drugs in the treatment
of MDR-TB.5 Over 40 adult patients with MDR/XDR-TB have
been reported who have received bedaquiline and delamanid
concurrently.84–87 The combination appears effective and well
tolerated, with few reports of-QTc intervals > 500 millisecond.
In some of these reports, patients were treated for greater than
the recommended 6 months and tolerated the medications
well.84,86,87 As these two drugs are often used in patients with
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
316 Management of Multidrug-Resistant Tuberculosis
few treatment options, extending the duration of use beyond
6 months is recommended in these settings.69
Group D3
Para-Aminosalicyclic Acid
PAS is a bacteriostatic agent that is highly specific for M.
tuberculosis. The drug is poorly tolerated, and, for this reason,
it should be used only when there are no other options.5
Resistant develops when mutations in thyA occur, but these
mutations account for only 6% of phenotypic resistance.43,88
Clinical trials demonstrated that monotherapy for 3 months
produced clinical improvement with efficacy similar to that
of streptomycin.89 The primary barrier to its use is the
ping
frequency of gastrointestinal side effects as well as- hyper-
sensitivity reactions. Hypothyroidism is common particu-
larly when combined with ethionamide.
PAS in the form of an enteric coated granule is widely used
because of its improved gastric tolerance and lower dosage
requirement.24 However, this formulation requires refrigera-
tion, which is not always available in developing countries.
Fortunately, an enteric coated granule has been developed that
does not require refrigeration. In contrast, PAS sodium lacks
the requirement for refrigerated storage. Neither formulation
is very active against M. tuberculosis, and given its high
frequency of intolerance, PAS was moved to group D3.7
Beta-Lactam Antibiotics
Mycobacterium tuberculosis is naturally resistant to β-lactam
antibiotics due to a class A β-lactamase, which hydrolyses
penicillins and cephalosporins. Resistance may be overcome
by two means: (1) inhibition of the β-lactamase adding
clavulanate acid or (2) the use of an antibiotic that is not a
substrate for it like the carbapenems. Ideally, both
approaches can be used by combining a carbapenem (imi-
penem or meropenem) with amoxicillin-clavulanic (because
clavulanic acid alone is not available in the market).5
β-lactam antibiotics such as amoxicillin plus clavulanic acid
and the carbapenems (imipenem, meropenem) have been
shown to have anti-TB activity that is enhanced in the presence
of clavulanate.45 In a mouse model, the combination of clavu-
lanate and imipenem/meropenem significantly improved sur-
vival, although they did not prevent bacterial growth as INH
did. Administration of amoxicillin using divided doses was
shown to have better EBA activity than single daily dosing, and
new formulations of amoxicillin/clavulanic (200/125 mg) may
be safely administered two to three times per day.45 In a small
clinical study of 10 patients with MDR-TB and multiple risk
factors for poor treatment outcomes, 1 g of imipenem given
twice daily was associated with sputum conversion in 8/10
patients: 7 patients remained culture-negative off therapy at
the end of the study.90 A case–control study of meropenem and
clavulanate plus linezolid-containing MDR-TB regimens was
reported to be associated with a smear conversion rate at
3 months of 87.5 versus 56% (p ¼ 0.02) in controls.91 More-
over, recently there have been several publications showing
the efficacy of imipenem, meropenem, and even ertapenem in
the treatment of the MDR and XDR-TB.92–95 Ertapenem has the
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
Daley, Caminero
advantage that it can be given intramuscularly and just once
daily. Based on these studies, it appears that the carbapenems
have significant activity against drug-resistant M. tuberculosis;
however, the high costs, lack of availability of oral formula-
tions, and variable market availability limit the usefulness
globally.
Conventional MDR-TB Regimen versus
Shorter Course Regimen
In the 2016 WHO guidelines, a 9- to 12-month shorter course
regimen was recommended for the first time.7 The shorter
course regimen can be used in children and adults who have
not previously been treated with SLDs and in whom resistance
to fluoroquinolones and second-line injectable agents has
been excluded or is considered highly unlikely. If the patient
does not meet recommended criteria reviewed below for
shorter course regimen, then an individualized regimen
should be used ideally with the addition of new or repurposed
drugs such as bedaquiline, delamanid, and/or linezolid.
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
The original exclusion criteria published by WHO are
listed here. However, since publication of the original criteria
in 2016, there have been subsequent discussions published
by WHO in December 2016 and recent publications with a
large number of MDR/RR-TB patients treated with this
shorter MDR/RR-TB regimen.96 Following is a reasonable
⑨
reappraisal of these exclusion criteria:
• Confirmed resistance or suspected ineffectiveness to a
medicine in the shorter MDR-TB regimen (except INH
resistance): in December 2016, the WHO published
responses to frequently asked questions noting that DST
of PZA was not a prerequisite to starting a shorter course
regimen due to the lack of available testing and that
clinicians may still decide to use the shorter course regi-
men even in the presence of PZA resistance. In addition, it
was not recommended to base treatment decisions on the
drug susceptibility results for EMB or any drug in the
regimen except for fluoroquinolones and second-line
injectables. A recent publication reported the experience
of using the shorter course regimen in nine African
countries. Among 1,006 MDR-TB patients, 81.6% were
treated successfully.96 Fluoroquinolone resistance was
the primary cause of failure, whereas resistance to PZA,
ethionamide, or EMB did not affect bacteriological out-
come. For second-line injectables, the success rate was
higher in susceptible than resistant strains, but the dif-
ference was not statistically significant. Therefore, based
on the WHO recommendations and the publication by
Trebucq et al,96 only resistance to the fluoroquinolones
-⑧
and/or SLD injectables should be an exclusion criterion for
the shorter MDR/RR-TB regimen.
• Exposure to more than one second-line medicines in the
shorter MDR-TB regimen for >1 month: this exclusion
criterion was included because of the fear of possible
acquired resistance to drugs used previously. But, as rea-
soned in the exclusion criterion 1, the confirmed resistance
to PZA, EMB, INH, and ethionamide/prothionamide has no
 Management of Multidrug-Resistant Tuberculosis
relevance in the final outcome of these regimens, whereas
confirmed resistance to either fluoroquinolones and/
or second-line injectables influences the outcomes of these
=
regimens. Therefore, this exclusion criterion should be
applied when fluoroquinolones and/or the second-line
injectables have been taken for a month or more, not for
the other drugs included in the shorter regimen.
• Intolerance to more than one medicine in the shorter
MDR-TB regimen or risk of toxicity (e.g., drug-drug inter-
actions): as described previously, this exclusions criterion
should relate to intolerance or a high risk of toxicity to
fluoroquinolones and/or second-line injectables only
(including drug–drug interactions).
• Pregnancy: this remains an appropriate exclusion criter-
ion because these patients were not included in previous
studies of the shorter course regimen and because of the
possible teratogenic effect of the injectable drugs.
• Extrapulmonary disease: This exclusion criterion was
included because the studies analyzing the efficacy of
the shorter MDR/RR-TB regimen did not include extra-
pulmonary TB. However, treatment of extrapulmonary
disease should be the same as that of pulmonary disease,
except perhaps in the setting of TB involving the central
nervous system (CNS).
• At least one medicine in the shorter MDR-TB regimen not
available in the program: this is also an appropriate
exclusion criterion because the shorter MDR/RR-TB regi-
mens should be offered as a package.
In conclusion, based on current information, resistance to
or intolerance of PZA, EMB, INH, or ethionamide/prothiona-
mide should not exclude someone from receiving the shorter
MDR/RR-TB regimen.
Conventional MDR-TB Regimen
€
Current WHO guidelines recommend atleast five effective
SLDs including PZA and four core SLDs: one from group A,
one from group B, and at least two from group C.7 Core drugs
include the later-generation fluoroquinolones and second-
line injectable agents, as well as other core drugs listed in
►Table 1. PZA is added routinely unless there is confirmed
resistance from reliable DST, or the drug is felt to be likely
resistant, or there is a significant risk of toxicity. The WHO
recommends that if the minimum effective medications
cannot be composed, then an agent from group D2 and other
agents from D3 may be added to achieve the goal of a total of
five drugs. It is recommended that the regimen be further
strengthened with high-dose INH and/or EMB.
The WHO 2011 update recommends an intensive phase of
8 months for most patients with modification based on the
patient’s response to therapy.26 This recommendation was
based on an analysis of relative risk for cure over successive
months of therapy. The adjusted relative risk for cure peaked
between 7.1 and 8.5 months, although this was not statisti-
cally different than 5.6 to 7 months or 8.6 to 20 months. In the
treatment of patients with newly diagnosed MDR-TB, a total
treatment duration of 20 months is suggested for most
Daley, Caminero 317
patients, but as with the intensive phase, this duration can
be modified based on the patient’s response to therapy.26
Using the same analytical approach, the peak in cure occurred
between 18.6 and 21.5 months for patients who had never
been treated for MDR-TB. In patients with a history of previous
MDR-TB treatment, no clear trend could be identified.
Shorter Course Regimens
The WHO recently updated their position on the use of shorter
course MDR-TB regimen following an expedited review of the
STREAM Stage 1 preliminary results and reiterated support of
the shorter course regimen.11 The initial experience with the
shorter course regimen was reported in 2010 from Bangla-
desh.50 In this study, patients with confirmed or suspected
MDR-TB were assigned to one of six standardized treatment
regimens administered in a serial fashion. The last regimen to
be studied, and the one with the best treatment outcomes,
included 4 months of kanamycin, clofazimine (100 mg/day for
those >50 kg), high-dose gatifloxacin (800 mg/day for those
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
>50 kg), EMB, high-dose INH (600 mg/day for those >50 kg),
PZA, and prothionamide. The intensive phase was prolonged
until smear conversion occurred or if the smear was still
positive at the end of 4 months of therapy. The continuation
phase was for five additional months (total of 9 months) and
included high-dose gatifloxacin, EMB, PZA, and clofazimine.
This regimen was associated with a treatment success of 87.8%,
death in 5.3%, default in 5.8%, failure in 0.5%, and relapse in
0.5%. Subsequent studies from Africa reported similarly good
treatment outcomes in programmatic settings.96
A recent systematic review and meta-analysis97 reported
outcomes from the aforementioned studies as well as data
from Uzbekistan and Swaziland. This review highlighted the
differences in the studies, such as which fluoroquinolone
was used and whether the usual dose or higher dose was
used, whether INH was used a usual or higher doses, and
whether prothionamide was given in the intensive phase or
throughout. Pooled treatment success was 97% and ranged
from 83.3% in Uzbekistan to 100% in Niger. When success was
compared with failure/relapse/death/lost to follow-up, the
proportion treated successfully dropped to 83%. However,
when compared with the historical success of conventional
regimens of 54% and programmatic data of 52%, the shorter
course regimen performed well. Lost to follow-up or not
evaluated dropped from more than 20% with conventional
and programmatic data to 5%. Risk factors for poor treatment
outcomes included fluoroquinolone resistance (odds ratio
[OR]: 46) followed by use of moxifloxacin rather than gati-
floxacin (OR: 9), PZA resistance (OR: 8), and no culture
conversion by 2 months (OR: 7). When death was compared
with survival, those who were HIV-infected were five times
more likely to die than HIV-negative patients.
Preliminary results were reported recently from the
STREAM 1 trial.11 Patients from seven sites were enrolled
between July 2012 and June 2015 and followed for 132 weeks
postrandomization. Patients were randomized to the shorter
course regimen (N ¼ 282) or longer conventional regimen
(N ¼ 142). A favorable outcome was marginally higher in the
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
 318 Management of Multidrug-Resistant Tuberculosis
shorter regimen than the conventional regimen, although it
was not statistically significantly different (80.6 vs. 78.1%;
p ¼ 0.6). When adjusted for the site of study and HIV
infection, there was 2.1% favor of the control arm; however,
the study did not satisfy the noninferiority threshold of 10%.
There were no statistically significant differences in other
outcomes such as patient retention, time to culture conver-
sion, relapse or mortality. However, in the subgroup of
patients with HIV infection, the number of observed deaths
was higher in the shorter course arm compared with the
control arm, but, again, the finding was not statistically
significant.
From a safety perspective, there were no statistical differ-
ences in the proportion of patients with AE of grade 3 to 5
severity; however, new onset prolongation of the QTc (Fri-
dericia corrected) interval to 500 millisecond or longer was
seen more commonly.11 Importantly, the shorter regimen
reduced the cost of treatment for the health system by an
average of US$2,879 per patient in Ethiopia (34% reduction)
and US$4,916 in South Africa (46% reduction). Based on these
findings, the WHO reiterated its conditional recommenda-
tion for the use of the shorter course regimen.
Surgery for MDR-TB
The WHO recommends that elective partial lung resection
(lobectomy or wedge resection) may be used with an appro-
priate treatment regimen in selected patients. Although
there are no randomized studies assessing the added benefit
of surgical resection over anti-TB chemotherapy alone, at
least two systematic reviews and data from an individual
patient meta-analysis have reported benefits in some
patients. One systematic review reported the results of 15
observational studies,98 and another review reported the
results of eight cohort studies of patients with MDR/XDR-TB
and an additional 18 retrospective case series.99 Treatment
success varied between 45 and 77%; the median postopera-
tive culture conversion was 93.5% (47–100%). Outcome data
from 26 cohort studies (18 surgical studies and 8 nonsurgical
studies) participating in the individual patients meta-ana-
lysis used for development of the WHO recommendations
showed a pooled treatment success of 84%, with failure in 6%,
relapse in 3%, death in 5%, and default in 3% of patients.7 In
the analysis, a statistically significant improvement in cure
and successful treatment was noted among patients who
received surgery. However, this benefit was primarily seen in
patients who had partial resection but not pneumonectomy.
Perioperative complications were reported in a median of
23% (0–39%) and perioperative mortality of 1.3% (0–5%). Risk
factors that have been identified to increase the risk of
i.EE?--
postoperative bronchopleural fistula include positive cul-
" tures at the time of surgery, polymicrobial infections, right
pneumonectomy, low FEV1, increased age, technique of
bronchial closure, and endobronchial disease.98–100
Based on these studies, it appears that surgery for MDR/
XDR-TB can provide additional treatment benefit ino selected
-
patients, but the procedure should only be performed by
⇐
experienced surgeons after the patient has been on appro-
Seminars in Respiratory and Critical Care Medicine Vol. 39
Daley, Caminero
priate therapy for several months, with the goal of achieving
smear and/or culture conversion preoperatively, if possible.
Prognosis appears to be better in those who underwent
①partial resection after culture conversion.7,101
-0
-
Special Situations
Extrapulmonary Disease
Current guidelines recommend that extrapulmonary TB is
the same as pulmonary disease. The American Thoracic
Society, Infectious Diseases Society of America, and Centers
for Disease Control and Prevention recommend extending
the duration of therapy for patients with CNS TB
(12 months).102 There are no specific recommendations for
the treatment of MDR-TB involving an extrapulmonary site.
- sea
However, if there is CNS involvement, it is critical to select
drugs that penetrate the blood–brain barrier. For example,
C-⑧
RIF, INH, PZA, prothionamide (ethionamide), cycloserine,
and later-generation fluoroquinolones have good penetra-
tion into the CSF. The aminoglycosides and polypeptides
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
penetrate into the CSF in the presence of inflammation;
therefore, their benefit is likely to be greatest during the
-
-
early phase of therapy. PAS and EMB have poor penetration
into the CSF.
Children
Treatment of MDR-TB in children is often delayed due to
difficulties in making a diagnosis due to the paucibacillary
nature of the disease and difficulty in obtaining adequate
diagnostic specimens. These delays can result in disease
progression, transmission to others, and death.70,103 Recom-
mendations regarding treatment are based primarily on small
observational studies and clinical/programmatic experience.
In general, the treatment regimens are built the same way for
children with MDR-TB as for adults; however, some experts
feel that in children with mild disease, second-line injectables
may not be needed to obtain successful treatment outcomes
and avoid toxicity. Data supporting this approach come from
the fact that in children with clinically diagnosed TB (as
opposed to bacteriologically confirmed), treatment success
was high and not significantly different in patients treated
with and without an injectable (group B) medication (93.5 vs.
98.1%).7 Drug selection in children is similar to that in adults,
but the SLDs are rarely produced in pediatric formulations or
appropriate tablet size. The optimal duration of therapy is not
known, but some experts believe that shorter durations are
possible in children with paucibacillary disease.104 However,
data supporting this approach are limited; therefore, most
-
experts recommend similar durations of therapy as adults. As
mentioned previously, children with MDR/RR-TB can receive
the shorter MDR-TB regimen.
A systematic review and meta-analysis reviewed treatment
outcomes for children with MDR-TB.105 Eight studies, which
reported outcomes on 315 patients, contributed to the data-
base. Average duration of treatment ranged from 6 to
34 months. The pooled estimate for treatment success (defined
as a composite of cure and completion) was 81.7%, with death
in 5.9%, and default in 6.2% of patients. Adverse reactions
No. 3/2018
 Management of Multidrug-Resistant Tuberculosis
occurred in 39.1% of the children, the most common of which
÷:-#
were nausea and vomiting followed by hearing loss, psychia-
tric effects, and hypothyroidism.
The WHO recently published interim policy guidelines
recommending that delamanid may be added to a WHO-
recommended longer treatment regimen in children and
adolescents (6–17 years) with MDR/RR-TB who are not
eligible for the shorter MDR-TB regimen.10 Data supporting
this recommendation are derived from phase I open-label
data and a subsequent open-label phase II extension study
demonstrating the safe administration of delamanid in this
population. There are no WHO recommendations for the use
of bedaquiline in children, although 27 children and adoles-
cents who received bedaquiline in programmatic settings did
well with no instance of drug withdrawal due to adverse
reactions.106
HIV-Infected Patients
Inadequate treatment of HIV-infected patients with MDR-TB
and XDR-TB has been associated with strikingly high mortality
rates.107 Thus, rapid identification of MDR-TB with initiation of
effective treatment regimens is critical. In addition, antiretro-
viral therapy (ART) is recommended for all HIV-infected
patients with drug-resistant TB irrespective of CD4 cell count,
and treatment should be initiated as soon as possible, generally
within the first 8 weeks of anti-TB therapy.26 Data supporting
this recommendation came from an analysis of 10 studies,
none of which were randomized trials. The analysis demon-
strated that there was a lower risk of death and a higher
likelihood of cure and resolution of TB signs and symptoms in
those patients using ART compared with those not using ART.
-
Recent studies in drug-susceptible TB demonstrated that ear-
lier initiation of ART (2 weeks after initiation of therapy) is
associated with improved survival, particularly in those
patients with CD4 cell counts less than 50 cells/mm3.108,109
Treatment of both MDR-TB and HIV is challenging due to
overlapping drug toxicities and potential drug interactions.
As mentioned previously, patients with HIV and MDR/RR-TB
can receive the shorter MDR-TB regimen.
Treatment Outcomes
Several systematic reviews and meta-analyses have reported
the pooled estimate of treatment success in patients with
MDR-TB and XDR-TB.110–113 The most recent review pub-
lished by Bastos et al110 included 74 studies and 17,494
patients. Pooled treatment success was 60% in MDR-TB and
26% in XDR-TB. The number of drugs, specific drug, and
duration of use were not associated with improved outcome.
MDR-TB patients receiving individualized treatments had
better outcome than those receiving standardized therapies
(64 vs. 52%; p < 0.001). The adverse drug reactions were
from 0.5% for EMB to 12.2% for PAS.
Predictors of poor outcomes were reported from five DOTS-
Plus projects in Estonia, Latvia, Philippines, Russia, and Peru
between 2000 and 2004.114 Out of 1,768 patients with MDR-TB,
treatment was successful in 65%, with deaths in 11%, default in
14%, and failed therapy in 7%. Independent predictors of death
Daley, Caminero 319
included age > 45 years, HIV infection, extrapulmonary dis-
ease, previous use of a fluoroquinolone, resistance to thioa-
mides, baseline positive smear, and no culture conversion by
the third month. Predictors of failure included cavitary disease,
resistance to any fluoroquinolone, resistance to any thioamide,
and no culture conversion by the third month. Default was
associated with unemployment, homelessness, imprisonment,
alcohol abuse, and baseline positive smear.
Monitoring for Treatment Response
The WHO recommends that treatment response be assessed
by -monthly sputum smear and culture rather than smear
microscopy alone.26 This strategy is the best strategy for
identifying failures earlier. Alternative modeling strategies
were examined based on cohorts of MDR-TB from Estonia,
Latvia, Philippines, Russia, and Peru from 2000 to 2004.115
Less than monthly monitoring results in delays in identifying
conversion, which would prolong intensive therapy, hospi-
talization, and respiratory isolation, increasing cost and
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
potential drug-related toxicity. Also, less frequent monitor-
ing would delay detection of treatment failure.
Monitoring for Adverse Reactions
Diagnosis of and monitoring for drug-related AEs are some of
the most important elements of case management in MDR-TB.
AEs due to SLDs are very common during the course of
treatment; the most common drug-related AEs and manage-
ment strategies are listed in ►Table 3. Among 818 patients in
the first five Green Light Committee projects, the most com-
mon AEs were nausea and vomiting (32.8%), diarrhea (21,1%),
arthralgias (16.4%), dizziness/vertigo (14.3%), and hearing
disturbances (12%).116 In a study from Latvia, 807 (79%) of
1,027 MDR-TB cases experienced at least one AE with a median
of three events per patient.117 These AEs resulted in a change in
the drug dose in 201 (20%) cases, whereas 661 (64%) of the
patients had at least one drug temporally or permanently
discontinued. Thus, AEs can negatively impact therapy result-
ing in default, morbidity, and even death.
Future Treatment Approaches
The current anti-TB drug development pipeline has at least
eight drugs in phase 2 and 3 trials.118 New compounds under
development include nitroimidazopyrans (pretomanid, for-
merly called PA-824), oxazolidinones (sutezolid, AZD-5847),
SQ109, benzothiazinones, and dinitrobenzamides.118,119
Pretomanid (PA-824) is a nitroimidazo-oxazine with a very
similar mechanism of action as that of delamanid. The combi-
nation of PA-824–moxifloxacin–PZA had the best mean 14-
day EBA (0.233 [SD 0.128]), significantly higher than that of
bedaquiline (0.061 [0.068]), bedaquiline-PZA (0.131 [0.102]),
bedaquiline-PA-824 (0.114 [0.050]), but not PA-824-PZA
(0.154 [0.040]), and comparable with that of standard treat-
ment (0·140 [0·094]).120 Preliminary data from a phase III trial
(NIX-TB) of an all-oral regimen (bedaquiline, pretomanid, and
linezolid) for the treatment of XDR-TB demonstrated that by
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018
320 Management of Multidrug-Resistant Tuberculosis Daley, Caminero

Table 3 Common adverse effects and management options

System Adverse effect Drugs Management options

Gastrointestinal Nausea and Eto/Pto, PAS, H, E, Z 1. Assess for dehydration; initiate hydration if indicated
vomiting 2. Initiate antiemetic therapy
3. Lower dose of suspected agent, if this can be done without compromising regimen,
rarely necessary
Gastritis PAS, Eto/Pto 1. H2-blockers, proton-pump inhibitors, or antacids (cannot be given with
fluoroquinolones)
2. Stop suspected agent(s) for short periods of time (e.g., 1–7 d)
3. Lower dose of suspected agent, if this can be done without compromising regimen
4. Discontinue suspected agent, if this can be done without compromising regimen
Diarrhea PAS, Eto/Pto, Clr, 1. Provide hydration in hospital, if severe
fluoroquinolones 2. Provide antidiarrheal agent as needed.
3. Search for other causes of diarrhea such as infections
4. Lower dose of suspected agent, if this can be done without compromising regimen
5. Discontinue suspected agent if this can be done without compromising regimen
Hepatitis Z, H, R, Eto/Pto, PAS, 1. Stop all therapy pending resolution of hepatitis
E, fluoroquinolones, 2. Eliminate other potential causes of hepatitis
bedaquiline 3. Consider suspending drugs, one at a time, with the most hepatotoxic agents first,
while monitoring liver function
Cardiac Prolonged QTc Fluoroquinolones, 1. Stop medications if QTc > 500 millisecond
bedaquiline,
delamanid

Downloaded by: Serials Unit - Periodicals. Copyrighted material.

Musculoskeletal Joint pain Z, fluoroquinolones 2. Initiate therapy with nonsteroidal anti-inflammatory drugs
3. Lower dose of suspected agent, if this can be done without compromising regimen
4. Discontinue suspected agent if this can be done without compromising regimen
Otological Hearing loss and S, Km, Am, Cm, Clr 1. Document hearing loss and compare with baseline audiometry if available
vestibular 2. Change parenteral treatment to capreomycin if isolate has documented suscept-
disturbances ibility to capreomycin
3. Decrease frequency and/or lower dose of suspected agent if this can be done
without compromising the regimen (consider administration three times per
week)
4. Discontinue suspected agent if this can be done without compromising regimen
Endocrinal Hypothyroidism PAS, Eto/Pto 1. Initiate levothyroxine therapy
Ophthalmic Optic neuritis E, Eto/Pto 1. Stop E
2. Refer patient to an ophthalmologist
Neurologic Seizures Cs, H, 1. Suspend suspected agent pending resolution of seizures
fluoroquinolones 2. Initiate anticonvulsant therapy (e.g., phenytoin, valproic acid)
3. Increase pyridoxine to maximum daily dose (200 mg/d)
4. Restart suspected agent or reinitiate suspected agent at lower dose, if essential to
the regimen
5. Discontinue suspected agent, if this can be done without compromising regimen
Peripheral Lzd, Cs, H, S, Km, Am, 1. Increase pyridoxine to maximum daily dose (200 mg/d)
neuropathy Cm, Eto/Pto, 2. Initiate therapy with tricyclic antidepressants such as amitriptyline; nonsteroidal
fluoroquinolones anti-inflammatory drugs or acetaminophen may help alleviate symptoms
3. Lower dose of suspected agent, if this can be done without compromising regimen
4. Discontinue suspected agent if this can be done without compromising regimen
Psychotic symptoms Cs, H, 1. Stop suspected agent for a short period of time (1–4 wk) while psychotic symptoms
fluoroquinolones, are brought under control
Eto/Pto 2. Initiate antipsychotic therapy
3. Lower dose of suspected agent, if this can be done without compromising regimen
4. Discontinue suspected agent, if this can be done without compromising regimen
Depression Cs, fluoroquinolones 1. Improve socioeconomic conditions
H, Eto/Pto 2. Offer group or individual counseling
3. Initiate antidepressant therapy
4. Lower dose of suspected agent, if this can be done without compromising the
regimen
5. Discontinue suspected agent, if this can be done without compromising regimen
Renal Nephrotoxicity S, Km, Am, Cm, Vm 1. Discontinue suspected agent
2. Consider using capreomycin if an aminoglycoside had not been the prior injectable
in regimen
3. Consider dosing two to three times a week if drug is essential to the regimen and
patient can tolerate (close monitoring of creatinine)
4. Adjust all antituberculosis medications according to the creatinine clearance
Electrolyte Cm, Vm, Km, Am, S 1. Check potassium
disturbances 2. If potassium is low, also check magnesium (and calcium if hypocalcemia is
(hypokalemia and suspected)
hypomagnesemia) 3. Replace electrolytes as needed

Source: Adapted from the World Health Organization Guidelines for the programmatic management of drug-resistant tuberculosis. Emergency update 2008.
Note: Drugs in bold type are more strongly associated with the adverse effect than drugs not in bold.

Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018

 Management of Multidrug-Resistant Tuberculosis
the end of 2016, 34 patients had completed 6 months of
therapy, and all culture converted with no relapses.121 Addi-
tional studies evaluating multiple different oral shorter course
regimens are currently underway.118 The STREAM study is
evaluating a new arm in the shorter MDR-TB regimen without
the injectable and with bedaquiline, levofloxacin, PZA, and
EMB for 9 months, and high-dose INH and prothionamide in
the last 4 months.
References
1 WHO. Global Tuberculosis Report 2017. Geneva: World Health
Organization; 2017
2 WHO. The End TB Strategy. Geneva: World Health Organization;
2014
3 Boehme CC, Nabeta P, Hillemann D, et al. Rapid molecular detec-
tion of tuberculosis and rifampin resistance. N Engl J Med 2010;
363(11):1005–1015
4 Hillemann D, Rüsch-Gerdes S, Richter E. Evaluation of the
GenoType MTBDRplus assay for rifampin and isoniazid suscept-
ibility testing of Mycobacterium tuberculosis strains and clinical
specimens. J Clin Microbiol 2007;45(08):2635–2640
5 Caminero JA, Scardigli A. Classification of antituberculosis drugs:
a new proposal based on the most recent evidence. Eur Respir J
2015;46(04):887–893
6 Caminero JA. Guidelines for clinical and operational manage-
ment of drug-resistant tuberculosis. Paris2013
7 WHO. Treatment Guidelines for Drug-Resistant Tuberculosis:
2016 Update. Geneva: World Health Organization; 2016
8 WHO. The Use of Bedaquiline in the Treatment of Multidrug-
Resistant Tuberculosis: Interim Policy Guidance. Geneva: World
Health Organization; 2013
9 WHO. The Use of Delamanid in the Treatment of Multidrug-
Resistant Tuberculosis: Interim Policy Guidance. Geneva: World
Health Organization; 2014
10 WHO. The Use of Delamanid in the Treatment of Multidrug-
Resistant Tuberculosis in Children and Adolescents: Interim
Policy Guidance. Geneva: World Health Organization; 2016
11 WHO. Position Statement on the Continued Use of the Shorter
MDR-TB Regimen following an Expedited Review of the STREAM
Stage 1 Preliminary Results. Geneva: World Health Organization;
2018
12 Migliori GB, Lange C, Girardi E, et al; SMIRA/TBNET Study Group.
Fluoroquinolones: are they essential to treat multidrug-resistant
tuberculosis? Eur Respir J 2008;31(04):904–905
13 Chan ED, Laurel V, Strand MJ, et al. Treatment and outcome
analysis of 205 patients with multidrug-resistant tuberculosis.
Am J Respir Crit Care Med 2004;169(10):1103–1109
14 Chiang CY, Enarson DA, Yu MC, et al. Outcome of pulmonary
multidrug-resistant tuberculosis: a 6-yr follow-up study. Eur
Respir J 2006;28(05):980–985
15 Zignol M, Dean AS, Alikhanova N, et al. Population-based resis-
tance of Mycobacterium tuberculosis isolates to pyrazinamide
and fluoroquinolones: results from a multicountry surveillance
project. Lancet Infect Dis 2016;16(10):1185–1192
16 Yew WW, Chan CK, Leung CC, et al. Comparative roles of
levofloxacin and ofloxacin in the treatment of multidrug-resis-
tant tuberculosis: preliminary results of a retrospective study
from Hong Kong. Chest 2003;124(04):1476–1481
17 Johnson JL, Hadad DJ, Boom WH, et al. Early and extended early
bactericidal activity of levofloxacin, gatifloxacin and moxiflox-
acin in pulmonary tuberculosis. Int J Tuberc Lung Dis 2006;10
(06):605–612
18 Peloquin CA, Hadad DJ, Molino LP, et al. Population pharmaco-
kinetics of levofloxacin, gatifloxacin, and moxifloxacin in adults
with pulmonary tuberculosis. Antimicrob Agents Chemother
2008;52(03):852–857
Daley, Caminero 321
19 Rustomjee R, Lienhardt C, Kanyok T, et al; Gatifloxacin for TB
(OFLOTUB) study team. A Phase II study of the sterilising
activities of ofloxacin, gatifloxacin and moxifloxacin in pulmon-
ary tuberculosis. Int J Tuberc Lung Dis 2008;12(02):128–138
20 Lee J, Lee CH, Kim DK, et al. Retrospective comparison of levo-
floxacin and moxifloxacin on multidrug-resistant tuberculosis
treatment outcomes. Korean J Intern Med 2011;26(02):153–159
21 Kam KM, Yip CW, Cheung TL, Tang HS, Leung OC, Chan MY.
Stepwise decrease in moxifloxacin susceptibility amongst clin-
ical isolates of multidrug-resistant Mycobacterium tuberculosis:
correlation with ofloxacin susceptibility. Microb Drug Resist
2006;12(01):7–11
22 Cheng AF, Yew WW, Chan EW, Chin ML, Hui MM, Chan RC.
Multiplex PCR amplimer conformation analysis for rapid detec-
tion of gyrA mutations in fluoroquinolone-resistant Mycobac-
terium tuberculosis clinical isolates. Antimicrob Agents
Chemother 2004;48(02):596–601
23 WHO. The Use of Molecular Line Probe Assays for the Detection
of Resistance to Second-Line Anti-Tuberculosis Drugs: Policy
Guidance. Geneva: World Health Organization; 2016
24 Caminero JA, Sotgiu G, Zumla A, Migliori GB. Best drug treatment
for multidrug-resistant and extensively drug-resistant tubercu-
losis. Lancet Infect Dis 2010;10(09):621–629
Downloaded by: Serials Unit - Periodicals. Copyrighted material.
25 Crowle AJ, Sbarbaro JA, Judson FN, Douvas GS, May MH. Inhibi-
tion by streptomycin of tubercle bacilli within cultured human
macrophages. Am Rev Respir Dis 1984;130(05):839–844
26 WHO. Guidelines for the Programmatic Management of Drug-
Resistant tuberculosis. Geneva: World Health Organization; 2011
27 Caminero JA; World Health Organization; American Thoracic
Society; British Thoracic Society. Treatment of multidrug-resis-
tant tuberculosis: evidence and controversies. Int J Tuberc Lung
Dis 2006;10(08):829–837
28 Heifets L, Lindholm-Levy P. Comparison of bactericidal activities
of streptomycin, amikacin, kanamycin, and capreomycin against
Mycobacterium avium and M. tuberculosis. Antimicrob Agents
Chemother 1989;33(08):1298–1301
29 Seddon JA, Godfrey-Faussett P, Jacobs K, Ebrahim A, Hesseling AC,
Schaaf HS. Hearing loss in patients on treatment for drug-
resistant tuberculosis. Eur Respir J 2012;40(05):1277–1286
30 Reuter A, Tisile P, von Delft D, et al. The devil we know: is the use
of injectable agents for the treatment of MDR-TB justified? Int J
Tuberc Lung Dis 2017;21(11):1114–1126
31 Georghiou SB, Magana M, Garfein RS, Catanzaro DG, Catanzaro A,
Rodwell TC. Evaluation of genetic mutations associated with Myco-
bacterium tuberculosis resistance to amikacin, kanamycin and
capreomycin: a systematic review. PLoS One 2012;7(03):e33275
32 Alangaden GJ, Kreiswirth BN, Aouad A, et al. Mechanism of
resistance to amikacin and kanamycin in Mycobacterium tuber-
culosis. Antimicrob Agents Chemother 1998;42(05):1295–1297
33 Krüüner A, Jureen P, Levina K, Ghebremichael S, Hoffner S.
Discordant resistance to kanamycin and amikacin in drug-resis-
tant Mycobacterium tuberculosis. Antimicrob Agents Chemother
2003;47(09):2971–2973
34 Maus CE, Plikaytis BB, Shinnick TM. Molecular analysis of cross-
resistance to capreomycin, kanamycin, amikacin, and viomycin
in Mycobacterium tuberculosis. Antimicrob Agents Chemother
2005;49(08):3192–3197
35 Sotgiu G, Centis R, D’Ambrosio L, et al. Efficacy, safety and
tolerability of linezolid containing regimens in treating MDR-
TB and XDR-TB: systematic review and meta-analysis. Eur Respir
J 2012;40(06):1430–1442
36 Cox H, Ford N. Linezolid for the treatment of complicated drug-
resistant tuberculosis: a systematic review and meta-analysis.
Int J Tuberc Lung Dis 2012;16(04):447–454
37 Koh WJ, Kwon OJ, Gwak H, et al. Daily 300 mg dose of linezolid for
the treatment of intractable multidrug-resistant and extensively
drug-resistant tuberculosis. J Antimicrob Chemother 2009;64(02):
388–391
Seminars in Respiratory and Critical Care Medicine Vol. 39 No. 3/2018