Immunology of Mycobacterial Disease
Item
- Title
-
Immunology of
Mycobacterial Disease - extracted text
-
d.
5
a. i -s. A v 1
(A t«u.
-—’ rlj
rv
(A
a
J
O.A-f •
U. A_
Antigen recognition
9'eiuzTL 2
79
Immune phenomena in mycobacterial disease the basic principles
Immunology of
mycobacterial disease
The immunological phenomena seen in mycobacterial disease consist as in
X“onl of recognition, response and react .on In the rs of th e
mens the invading mycobacteria arc recognized as being, foreign . Ln tn
second step the necessary defence mechanisms are alerted and fruited
while in the third the actual struggle between the mycobacterium and the ho
takes place.
Antigen recognition
p.d.og.n,. In ISO c^f0[h(ep™wre ”
“ ,1*
disease control.
PXant to an understanding of the pathogenesis of
^ZldtXX nature and^hcance of t—
“ “■
deficiency v‘rus (HIV)
Mycobacteria, in col
intracellular parasites, owe their virumacrophages. Protective immune reac-
iiSSHESSEESSSfe
cen? while the latter includes suppressor and cytotoxic cells. The association
of CD antigens and functional activity is, in fact, not so dear-cut as‘ ong,na *
Sough? as cytotoxic CD4+ cells have been described (Ottenhoff er al., 1988,
Fly™V^
of T-cells are largely due to the
c Jokines that they secrete. A number of well-charactenzed^cytokinespromany attempts
they
do, not appear to play a major role * »
1 .
types of protective
esT5ncjtyne, which until
to protection, is the abllltY of certaim ce
intracellular bacillary
macrophages and othei‘ c^ls g - Tea^myCobacteria, thus enabling them
These are recognized by the nature of the cytokines that they, secrete (Table
I n and ire termed THLandJB2. There is also evidence that CD8_+ (supr so^ ") Ts likewise mature along these^athways so t e terms
Type 1 and Type 2 are often used instead of TH1 and TH2. This division
Table 5.1 Cytokines produced by TH1 and TH2 T-cells
IT thbera 19® "ft of etnphSsimportant one as cell
a^vitS thought to be associated only w.th tmtnunopathology and progression of disease.
Cytokine
Gamma interferon
lnterleukin-2
lnterleukin-4
lnterleukin-5
lnterleukin-6
Interleukin-10
TH1
TH2
80
The immune response: macrophage activation
Immunology of mycobacterial disease
two maturation types is of great relevance to the nature of the immune
response following infection by a mycobacterium, as described below.
To complicate the matter further, feelper T-cells are also divisible according
to differences in the structure of the receptor molecules that bind to antigen.
These receptors consist of two protein chains which, in most cells, are of types
termed alpha and beta but a minority of cells have gamma and delta chains.
The latter, termed gamma-delta (y-S) cells are usually CD4-/CD8- but a
minority are CD8+.
.
Each lymphocyte bears a receptor that binds to just one of the thousands oi
possible antigenic determinants or epitopes. Accordingly, there must be many
thousands of subpopulations of lymphocytes, each specific lor just one epi
tope An important stage of the immune response is that of clonal expansion,
in which small numbers of antigen-specific lymphocytes proliferate to form
clones of cells of sufficient numbers to mediate an effective immune response.
Before an antigen is able to induce such clonal proliferation it must be pre
sented to the lymphocytes in a special way. This is the task of the antigen pre
senting cells (APCs) which include cells of the monocyte/macrophage series,
dendritic cells of lymph nodes and scattered lymphoid tissues and the
Langerhans cells of the dermis. Other cells may, under certain circumstances,
also serve as APCs. The process of antigen presentation is shown in Fig. 5.1.
The APCs engulf mycobacteria which contain a multitude of epitopes. Some
of these epitopes are actively secreted (see Chapter 2) and some are somatic,
and only released by bacteria that are digested, to a varying extent, by
enzymes within lysosomes.
.
Mycobacterial epitopes arc presented on the surface of the APC in close
relation to host molecules which arc products of the major histocompatibility
(CD4J
(cD?)
CDS
CDS
CDS
(cd?)
(cd?)
Antigen processing cell
Fig. 5.1 Antigen presentation. A complex antigen is phagocytosed and degraded by the
antigen processing cell (APC). Individual epitopes derived from antigens within
phagosones are presented on the cell membrane by the Class II MHC (HLA-D) molecules
to antigen-specific CD4 T-cells while those derived from antigen in the cytoplasm are
presented by the Class I (HLA-A and HLA-B) molecules to CDS T-cells (T). In each case,
this process activates the T-cells which then proliferate to form clones
81
complex (MHC). There are two types of such molecules: Class I, coded for by
the HLA-A and HLA-B genes, and Class II, coded for by HLA-D genes. The
Class II molecules are principally found on cells with specific antigen pre
senting functions and activate the CD4+ helper/inducer T-cells. Class I mole
cules are present on all cells and activate the CD8+ suppressor/cytotoxic
T-cells Binding of antigen-specific lymphocytes to the epitope/HLA complex
causes a signal to be delivered by the APC to the lymphocyte, inducing the
latter to secrete interleukin-2 (IL-2) which mediates lymphocyte division and
clonal expansion. Epitopes from pathogens within phagosomes are presented
by MHC Class II molecules while those from pathogens that lie freely in the
cytoplasm of the cell arc presented by the Class 1 molecules. This difference
in presentation enables the immune system to ‘decide’ whether to enhance the
microbicidal power of the cell or to destroy it and thereby liberate its contents.
The immune response: macrophage activation
The next cells to enter the scenario are the macrophages which belong to the
same cell lineage as the blood-borne monocytes. Macrophages are phagocytic
cells but, unlike the other major class of phagocytes, the polymorphonuclear
leukocytes, they are long-lived cells which settle in a given tissue or organ and
adapt to the local environment. Thus, alveolar macrophages, osteoclasts,
Kupffer cells of the liver and Schwann cells of the nerves are all specialized
macrophages. Further, macrophages do not express their full antimicrobial
potential unless they are ‘activated’. Such activation is mediated by gamma
interferon (IFN-y) which is secreted by the clonally expanded population of
CD4+ helper T-cells (Rook et al., 1986).
In man, there is an additional step in macrophage activation involving vita
min D. Macrophages activated by IFN-y produce a hydroxylase enzyme
which converts inactive vitamin D into its active metabolite calcitnol which
further activates the macrophage (Fig. 5.2; Rook, 1986). This may well
explain the success of vitamin D therapy in the treatment of lupus vulgans in
the pre-chemotherapy era.
.
The activated macrophage differs from its resting counterpart in several
respects. The cell membrane is much more motile - a phenomenon termed
membrane ruffling. Random migration, glass adherence and the ability of the
cell to phagocytose and kill micro-organisms are all increased.
In addition to its microbicidal activity, the macrophage synthesizes and
secretes many important compounds that affect the pathogenesis of mycobac
terial disease. These include some of the acute-phase reactant proteins,
vasoactive peptides, and proteases that liquefy necrotic tissue and thereby
contribute towards the formation of the tuberculosis cavity. In addition,
macrophages produce a cytokine termed tumour necrosis factor (TNF-a;
Flesch and Kaufmann, 1990). This cytokine has a protective function and
treatment of mice infected with Bacille Calmette-Guerin (BCG) or
Mycobacterium tuberculosis with antibody to TNF-a leads to rapid progres
sion of disease (Rook and Bloom, 1994). On the other hand, TNF-a is an
important mediator of tissue-necrotizing immunopathology as desenbed on
page 89. Tumour necrosis factor-alpha is also known as cachectin and its
i lie iiiiiiiiuie reuctiuii: events hu/.u. <,.u n.L.u.^
I
82 Immunology of mycobacterial disease
TH
o >
TNF-y
W) o
th)
O
TH '(th) (th)
25 OHD3
O o X
O
o O
\ (M<? / o
I ox—
/o ° o /
1,25 OH2D3
I
TH
ARC
<>5
>2 00°° e'
SssS
Thl more aS 1.25-OHJ), metaWte which causes further macrophage
activation. Data from Rook (1986)
systemic release is responsible for the extreme wasting associated with
advanced tuberculosis (Beutler et al., 1985).
fnrmation of the
An important characteristic of chrome inlections is the format o of the
oranuloma which is a compact cordon-like aggregate, many cells thick, of
activated macrophages around the site of infection. Macrophages in granulonfas am termed epithelioid- cells, from their '™n>tologmal similanty o
cpilh,
i'ial ce”t(S;^, ^dy XS1 wten?Mr
mb'crcies by Hippocrates who likened them' to miniature tubers of plants;
hcncc the name tuberculosis.
The immune reaction: events within the
macrophage
In order to survive within a macrophage, a bacterium must be able to resist
destruction by the wide range of non-oxygen-dependent and oxy?endtpe^ent killing mechanisms of the infected cell. The former mechamsms
include enzymes such as lysozyme, lipases and phosphatases and thelatte!
include generation of reactive oxygen intermediates such as superoxtde radi _
cals hydrogen peroxide and hypochlorite ions. In the mouse, nitric oxide ai d
relaied reactive nitrogen intermediates> (RNIs) generated1 frorr^-^ni p
vide powerful means of killing mycobacteria (Chan er al 19.)2 bul^it isi^t
certain whether human macrophages generate RNIs (Rook and Bloom, 199J.
as
.1^
Fig. 5.3 A tuberculous granuloma showing whorls of epithelioid cells, some giant cells and
central necrosis
The process of phagocytosis is shown in Fig. 5.4. An engulfed bacterium
lies within a vesicle formed by invagination of the surface membrane. This
vesicle the phagosome, then fuses with the lysosomes which contain the bac
tericidal agents referred to above. There are three main strategies by which
bacteria or other pathogens survive within the phagosome. First, phagosome/lysosome fusion may be inhibited. Secondly, the pathogen may covei
itself with a protective layer that absorbs or neutralizes the bactericidal agents
or, thirdly, it may escape from the vesicle and lie freely in the cytoplasm of the
CelThe extent to which mycobacteria use these strategies is controversial.
Mycobacterium tuberculosis, in common with the protozoal parasite
Toxoplasma gondii, inhibits phagosome-lysosome fusion but neither the
mechanism nor the significance of this activity to intracellular survival are
clearly understood (Draper, 1981). Mycobacterial pathogens also appear to
survive the effects of exposure to reactive oxygen intermediates (ROI) on
84
Immunology of mycobacterial disease
B
C
E
0
D
Fig. 5.4 Phagocytosis. A: Bacilli are engulfed by the cell membrane. B: The bacilli lie in a
membrane vesicle - the phagosome - which fuses with lysosomes containing bactericidal
substances. C: These substances destroy the bacilli. Mycobacteria avoid such destruction
by - D: inhibiting phagosome/lysosome fusion; E: a thick capsule-like outer protective
layer; and F: escape from the phagosome to lie freely within the cytoplasm
account of the thick outer layer of mycosides which, on electron microscopy,
appears as an electron transparent zone surrounding the bacilli. Lipoarabinomannan (LAM), which is present in all mycobacteria, and phenolicglycolipid-I (PGL-1) in M. leprae also protect against ROI (Chan et al.,
1991). In addition, mycobacteria secrete the enzyme superoxide dismutase
(SOD) which also protects against ROI (see Chapter 2, page 15).
Mycobacteria are able to escape from the phagosome and replicate in the
cytoplasm (McDonough et al., 1993). This appears to occur when the cell
becomes immunologically effete and unable to control intracellular growth or
when mycobacteria enter cells other than macrophages. Epitopes from the
cytoplasmic location are presented to T-cells by the MHC Class I molecules
which, as mentioned above, are present on all cells and facilitate cell lysis by
CD4+ and CD8+ cytotoxic T-cells.
The early immunological events after infection
The events summarized above underlie the pathogenesis of primary pul
monary tuberculosis. The infectious particle is a small droplet of cough spray,
about 5 pm in diameter, containing a few tubercle bacilli. After inhalation,
this lodges in an alveolus, usually near the periphery of the lung, and the
bacilli are engulfed by alveolar macrophages. If the macrophage cannot con
trol bacillary growth, it is killed with release of the bacilli which initiate an
inflammatory response which, in turn, attracts blood-borne phagocytes and
other white cells, including natural killer cells and y-5 T-cells. The latter cells
recognize certain, as yet not clearly defined, mycobacterial components and
cause the formation of a low-turnover granuloma of the ‘foreign body’ type.
This to some extent, limits the progression of the disease process before the
Delayed hypersensitivity and the Koch phenomenon
85
development of the specific immunity mediated by CD4+ T-cells. There is
evidence that early recognition of mycobacteria by y-8 cells is required tor
subsequent activation of antigen-specific a-p CD4+ T-cells (Kaufmann et al.,
1993). Mycobacterial adjuvants (see page 105) induce granuloma formation
and there is evidence that this may occur by a T-cell-independent pathway
(Bancroft et al., 1991).
Some tubercle bacilli in the primary focus are transported, probably within
phagocytic cells, to the local lymph nodes, where antigen is processed and
presented to antigen-specific lymphocytes as described above. Some bacilli
are carried further afield in the lymphatics and blood stream and are respon
sible for the serious non-pulmonary manifestations of primary tuberculosis
described in Chapter 8.
When the specific immune response has developed, the low cellular
turnover ‘foreign body’ type granuloma gives way to the higLfurnoyer granu^
loma of immunogenic origin. In tuberculosis the granuloma consists of a cen
tral area of cheese-like necrosis, or caseation, surrounded by epithelioid cells
and with lymphocytes in the outer zone. Some epithelioid cells fuse to form
multinucleate cells (Langhans’ giant cells). In many cases this protective
immune response is sufficient to arrest the disease and destroy most or the
mycobacteria. Collagen is then laid down by fibroblasts and the foci heal by
scarring.
.
...
.
' Similar resolution of early disease also occurs in many cases of leprosy and
other mycobacterial infections, although it has not been possible to examine
the sequence of events in the same detail as in pulmonary tuberculosis.
Protective cell-mediated reactions as described above are not the only
immune phenomena associated with mycobacterial disease. Others, such as
immunosuppression and delayed hypersensitivity, also occur and make the
subject much more complex. Indeed these other reactions are of great rele
vance to the pathogenesis of the diseases.
Delayed hypersensitivity and the Koch
phenomenon
A hypersensitivity reaction is defined as one which causes tissue damage.
Four main types are recognized; namely, anaphylactic (Type I), antibody
dependent cytotoxic (Type II), immune complex-mediated (Type HI), and
delayed (Type IV). The first three involve antibody but delayed hypersensi
tivity is ‘cell-mediated’. The tuberculin reaction is often cited as the classical
example of the delayed type hypersensitivity (DTH).or l\pe IV reaction but,
in fact there are a number of different DTH reactions, some of which cause
tissue necrosis and others which do not. Both types occur in mycobacterial
disease and appear to bear quite different relations to protective cell-mediated
immune reactions. For this reason, the relation between DTH and CMI has
long been the subject of controversy.
.
In order to understand DTH it is necessary to look back to the original
studies of Koch (1891) which were carried out in an attempt to discover a
cure for tuberculosis. Koch inoculated tubercle bacilli into the flanks of
86
Immunology of mycobacterial disease
guinea-pigs and observed the ensuing events. After a week or two a small,
firm nodule developed at the inoculation site and subsequently ulcerated.
Viable tubercle bacilli were isolated from the ulcer which remained open
until the animal died. About a month after inoculation local lymph nodes
were enlarged and disseminated disease developed, leading to death three or
four months later. When Koch gave a similar inoculation of bacilli into the
opposite flank one month after the original infecting dose a quite different
lesion developed. After a day or two the skin at the second inoculation site
became black and necrotic and then sloughed off leaving a shallow ulcer
from which no bacilli could be isolated and which soon healed. Koch then
found that he could elicit a similar reaction, subsequently termed the_Koch
phenomenon, by injecting either killed tubercle bacilli or a heat-concentrated
filtrate of the medium in which the bacilli had been grown, a preparation
terThe^mportanr^ note is that, although this reaction clearly led to the
elimination of bacilli inoculated into the skin, the animals were nevertheless
dyingofjystemic tuberculosis. This implies that either the rcacti^LdoeEnot
occur4n-lhe presence of tubercle bacilli in the deep tissues or. 11 it does it is
either non-protective or positively harmful. In fact, Koch himself put the
matter to the test by administering Old Tuberculin systemically to patients
with tuberculosis. Although there were a few remarkable cures in patients
with disease of the skin or larynx there was little or no effect in those with
deep lesions. Indeed, therapy in some patients led to a worsening of pul
monary tuberculosis and some developed ‘tuberculin shock which in a few
cases proved fatal. Thus it appears that if the necrotic Koch phenomenon
occurs on the surface, the bacilli-laden tissue easily s oughs off. If, on the
other hand, it occurs in the lung or other internal organ, the bacilli and necrotic
A
tlSKoch’Twork on tuberculin would probably have been forgotten had it not
been for the extensive studies of the Austrian_phY.sician Cleip^_s_yon Pirquej
who showed that dermal reactivity to a small quantity of tubercuhii w_as
indicative of past infection by the tubercle bacillus and thus of great epidemi
ological value (von Pirquet, 1907).
Post-primary tuberculosis and the Koch
phenomenon
Conversion to tuberculin positivity in humans occurs about six to eightweeks
after the initial infection. By then the primary lesion is, in many cases, well
contained. Post-primary disease occurs months, year_s_or even JecadesJater.
and is the result of either reactivation of old, latent, foci of disease.^ o
exogenous reinfection (see Chapter 8).
.
Post-primary lesions which, for unknown reasons, often occur in the apica
regions of the lungs, are characterized by extensive tissue necrosis. This was
observed by Sylvius in 1680 who wrote T may clearly communicate that I
saw on many occasions glandulous tubercles in the lungs which sometimes
contained various forms of pus as a section showed (Sylvius 1680).
Protective immunity and immunopathology in tuberculosis
87
The extensive necrosis results in large, caseating, tumour-like lesions
termed tuberculomas being formed. As in primary lesions, these lesions are
acidic and anoxic, and they contain tissue-derived free fatty acids. Thus the
number of viable tubercle bacilli within them is low.
,
The caseous material in the post-primary lesions is softened or liquefied by
proteases liberated by macrophages and, if the lesion erodes into a bronchus,
the softened contents are coughed out leaving a cavity. In distinct contrast to
the closed lesion, the cavity is well-oxygenated and becomes an ideal breed
ing ground for tubercle bacilli. Thus the cavity wall contains millions of freely
replicating bacilli which are behaving more like saprophytes than primary
pathogens. In addition, large numbers of bacilli enter the sputum, rendering
the palient open or infectious. Some cavities close spontaneously due to con
traction of fibrous scar tissue, the anoxic conditions return and the number of
bacilli decrease. In the pre-chemotherapy era, closure of cavities was achieved
therapeutically by inducing lung collapse by artificial pneumothorax or by sur
gical resection of parts of the chest wall - an operation known as thoracoplasty.
0 It is therefore evident that immunological reactivity of the tissue necrotiziiTg_type has a profound.effect on the pathogenesis of the disease and is an
important factor in determining infectivity.
Tubercle bacilli escaping from cavities may cause secondary lesions in the
lower lobes of the lung, in the upper respiratory tract and, if swallowed, in the
alimentary tract (Chapter 8, page 166). On the other hand, lymphatic and I
haematogenous dissemination of disease is, in contrast to primary tubercu- J
losis, uncommon/Thrs is probably due to necrosis of the draining lymphatics
an^capitraries by the DTH reaction. An important component of the protec
tion afforded by BCG vaccine is the preyentioiLoLserioiis lorms of pnniary
tuberculosis that resulTfrom haematogenous dissemination, and this may be
due to the induction of DTH (Ladefoged et al., 1976).
Protective immunity and immunopathology in
tuberculosis
For many years the relationship between protective cell-mediated immunity
(CMI) and DTH was the subject of much controversy. Some workers claimed
that these are essentially similar phenomena but differ in degree while others
argued that they are separate and distinct reactions.
^During the two decades following the introduction of the tuberculin test as
a diagnostic and epidemiological tool by von Pirquet, it was widely asserted
that tuberculin reactivity was a sign and measure of immunity. This view was
seriously challenged by Rich and McCordock (1929) and many other workers
in the ensuing decades (reviewed by Bothamley and Grange, 1991). There is
now considerable evidence that small tuberculin reactionjcorrelatewit h protection but that larger ones are indicative of tissue-desiroying-hypgisensj.tjvity.
(Fine, 1994). While some have argued that the difference between a protective
CMI reaction and a necrotic DTH reaction is one of degree, evidence that the
differences are qualitatrvTrather than merely quantitative has steadily accu
mulated.
I
88
The role ofTNF in the immunopathogenesis of mycobacterial disease
89
Immunology of mycobacterial disease
contact. Thus if BCG induces or boosts protective immunity it will appear
effective but if it boosts_predetermined necrotic Koch-type reactivity it will
A major step forward in the resolution of this controversy was taken by
Rook and Stanford (1979). In a detailed study of mice experimental y
infected with various mycobacteria, these workers observed that a reaction to
tuberculin peaking at about 20 hours after skin testing appeared around 10
days after infection, while a reaction peaking at about 40 hours appeared a
month or so after infection. The former reaction, which was non-necrotizing,
resembled that demonstrable in mice infected with Listeria monocytogenes
while the latter appeared to be the murine equivalent of the necrotic Koch
phenomenon in guinea-pigs. These reactions were therefore termed the
Listeria-type and the Koch-type reactions, respectively. Interestingly, the less
mouse-virulent daughter strains of BCG and certain environmental mycobac
teria not known to cause disease preferentially elicited the non-neciotic
(Listeria-type) reaction, while the more virulent of the BCG strains and othei
mycobacteria that are pathogenic in the mouse elicited the Koch-type ieaction. Furthermore, the induction of one type of reaction appeared to block the
subsequent induction of the other. Thus a predisposition jo respond to subse
quent mycobacterial challenges with the ‘Listeria-type rc^onsc ^u.ld .
induced by incorporating a rapidly growing non-pathogen, M. vaccae in
animals’ drinking water. These findings support the concept that imipgne
reactivity in. man is determined by the nature of the immunological y effeyfc^ct with mycobacteria early in life - a phenomenon termed onginal
mycobacterial sin’ by Abrahams_(197QX.--The practical relevance of these findings came from the observation that the
efficacy of BCG vaccination varies greatly from region to region (see page
97). One of the most plausible of many explanations of this variation is that
the efficacy of this vaccine is predetermined by previous exposure to myco
bacteria in the environment. In studies involving guinea-pigs, Palmer and
Lons (1969) found that such exposure afforded some protection against infec
tion by M. tuberculosis. Vaccination with BCG increased this immunity but
never to a level above that induced by BCG alone. It was therefore concll*^d
that the full measurable effect of BCG is seen in populations not previous y
immunized by contact with environmental mycobacteria, but that clscwhe
the observed effect is less as the population already has some naturally
dC I^couldTl^refore be postulated that in regions where BCG is ineffective the
population would have received sufficient ‘natural vaccmahon to_induce
maximum immunity before vaccination. Indeed, in this snuation,^BCQ^P
even push vaccinated persons from a protective to a hypersensi ive state,
thereby reducing their resistance to disease. If this theory is correct, the vac
cine should confer protection if given to neonates or young children befoie
they experience a significant exposure to mycobacteria in the environment.
An akemative explanation of the apparent differences in the efficacy of
BCG was advanced by Stanford and his colleagues on the basis of thei
description of the non-necrotizing and necrotizing tuberculin reactions
described above and the finding that some forms of ^munologica •v
contact with mycobacteria induce the former reaction while others induce the
latter (Stanford et al., 1981b). While BCG induces a non-necrotic, protective
response in those not previously sensitized by mycobacteria, in otheis i
hJU whatever pattern of reactivity has been ‘imprinted by environmental
nOATPXePrvTdorof fundamental importance in clarifying the elation
between CMI and DTH was the demonstration that the two reactions could be
adoptively transferred separately in mice by different T-cell clones (Orme and-Collins, 1984). Subsequently a lot of light was shed on the nature of this dis
sociation at the cellular level and on the way in which environmental myco
bacteria predetermine subsequent immune reactivity by the demonstration of
the two T-cell maturation pathways', TH1 and TH2 (see above, page 79).
Bretscher has shown that contact with mycobacterial antigen, even if in too
small an amount to induce detectable immune responses, may imprint the
immune system with a tendency to respond to subsequent contact withi myco
bacteria with either a TH 1 - or TH2-mediated reaction (Bretscher, 1992). Thus,
though containing numerous different epitopes, a complex antigen such as a
mycobacterium induces a remarkably unified pattern of immune responses, a
phenomenon which Bretscher has termed ‘coherence’.
The balance between TH1 and TH2 achieved by immunization, and thus
probably by natural infection, is determined by a bal^c^n^e.t'v^.n van(^.s
steroid hormones (Daynes et al., 1991; Rook et al 1993, 1994).
coids such as cortisol promote TH2 maturation while dehydroepiandrosterone
(DHEA) opposes this effect and promotes TH1 maturation Anjncreas?_ip
cortisol levels relative to PHEA^occurs-imXuberculosis and AIDS and raises
the possibility bFTectifying this imbalance by an immunotherapeutic agent
with TH1 adjuvant properties (Rook et al., 1994).
I
Cytotoxic cells and protective immunity
pSog^nsprZes™
mechanisms (Fig. 5.5). As mentioned above, there is now evidence that cyt^
toxic cells recognizing mycobacterial antigen are essential for protective,
immunity (Boom el al , 1991; Flynn el al., 1992). There are also ess specific
mechanisms for lysing infected cells, including natusal killer ceiK cytiiloxic
T-cells that recognize proteins expressed on stressed cells, and TNh these
additional mechanisms are discussed below.
The role of tumour necrosis factor in the
immunopathogenesis of mycobacterial disease
I
As discussed above (see page 81), TNF-a makes a major contribution to
protective immune responses in mycobacterial disease by inking-and
taining granulomas. Paradoxically, though, ins_alsp_a key factor in use
destroying reactionsJeading to progression of disease The reason for, this
that macrophages and other cells infected with virulent strains of M. tuberci
losis are rendered exquisitely sensitive to killing by TNF-a (Filley et al..
Tuberculin and tuberculin reactivity
90
Immunology of mycobacterial disease
CD8
0
CD4
@@
(TH?)
(jHl)
A' '3
L
0®
y
I
CDS
91
(TH?)
@
y
Cell
\W XA
O
M0
(M0 )
M. Tuberculosis^
*
I
r'
LAM
O
o
TNF-a
Calcitriol
M0
Fia 5 5 The role of cytotoxic cells in protective immunity. The cytotoxic= (usually_C^8^
T-cel?lyses a cell containing many mycobacteria in the cyl°P'as^ en^1'n9
b=
engulfed by an immunologically effective macrophage activated by CD4+ T-cells. M(|)
macrophage
ans, m
in addition to ceii-meaiamu
cell-mediated cyio
cytotoxi1992) This may well provide a means,
umcity, by which tubercle bacilli are released from cells unable to control the
'"'l^addition, infection by mycobacteria elicits a more generalized effect of
TNF-a on tissues. This is analogous to the Shjvart_z_m^_phenomenon in
which injection of a Gram-negative endotoxin into the skin primes the sttt1 for
a necrotic reaction when the same endotoxin is given intravenously 24 hou s
later The Koch phenomenon is somewhat similar because, as ori0inally
demonstrated by Koch (1891), systemic injection ofluberculin causes neciosis around tuberculous foci, with their elimination if they are superficiak
, Ti^our necrosis factor-a is involved in the Koch' Pheno™e?on?;^^
is accompanied by a massive systemic release of this cytokine (Rook and Al
ATyhehklc9h1phenomenon elicited by mycobacteria differs in one important
aspect from the Shwartzman phenomenon induced by Gram-negative> endo
toxin in that the former is T-cell.dependent (Al Attiyah et al., 1992) !n addi
tion priming for necrosis requires cytokines from TH2, oi a mixture of TH
and TH2, T-cells. Cytokines from a pure TH1 T-cell
d°qJSi\PIThis
the tissues for necrosis (Fig. 5.6; Hernandez-Pando and Rook, 1994) This
finding is of crucial importance as it explains how the T-cell ma^.rat^n P|:
and the environmental and other factors determining the type, cntically
the nature of immune_responsiyeness_to^cliaUeiige^ bypathogenic my^
bacterialnd thereby the outcome of such a challenge. It also provided
(and granuloma formation). TH2 cells produce a substance’
renders cehs and
M<J> = macrophage
rationale for the development of successful immunotherapy for mycobacterial
disease as described in Chapter 10.
Tuberculin and tuberculin reactivity
Koch’s Old Tuberculin was a filtrate ofjlhrolh culture of Mjuberwlosu con
centratedby evaporation in a heated water bath. It contained various impunties derived from the medium and tended to induce non-specific inflammatory
reactions. To overcome this problem Seibert (19341 attempted to harvest the
tuberculoproteins by precipitation with acetone and ammonium ^phate- The
resulting
resulting preparation
preparation was termed purified_prQte.in denvative of lubercu in
(PPD) and has been used widely since. Despite the name, PPD is not pure
^Koch also produced New Tuberculins by grinding mycobactena thereby
releasing their cytoplasmic antigens. This method of producing mycobactenal
skin test reagents was reintroduced by Stanford and his colleagues. These
reagents are now prepared by harvesting mycobacteria during
growing
phase from non-antigenic media, washing them thoroughly, disrupting the cell
Categories oj tuberculin reactors
92
vj
Immunology of mycobacterial disease
mass in an ultrasonicator, sterilizing the cytoplasm by repealed membrane: fil
tration, and diluting it to a suitable protein concentration. The firs. ot the sc
reagents was Burulin, prepared for studies on Buruh ulcer in Ahica (Stanford
et al 1975), but subsequently they were prepared from many mycobacterial
species (Editorial, 1984). NewTuberculins are relatively much richer in the
“"'-I J-
^TTtX^ypes^fTintest reagents prepared from M. leprae
leprosins) and the reactions elicited by them are described in Chapter 7 (see
PaA positive tuberculijneaction usually manifests as an area of induration
which reaches a maximum after 48 or 72Jtours. Eryfepa also occurs and
may be much more extensive than the induration; but it is by convention
ignored as it is difficult to see and measure in dark-skinned peoples, ^onp
patients with active tuberculosis have visible reactions as_ead5Las^i^^
\ after testing (Kardjito and Grange, 1982). This early reaction is particularly
'prominent amongst healthy and radiologically clear persons who are occupa’ tionally exposed to patients with tuberculosis, suggesting that it is associated
wlth proteSon (El Ansary and Grange, 1984; Grange- et al 1986) Studies of
punch biopsies of the early reaction show that it is a typical DTH reactionot
^The ^aTe c^onent of thf tuberculin reaction may also be caused by quali
tatively different reactions. Stanford and Lerna (1983) observed that some hour tuberculin reactions in humans are purple co loured, ,ndJa^
demarcated and tender, while others are pink, soft, ill-defined and much css
tender. It has been postulated that these reactions correspond respectrvely
the necrotizing (Koch-type) and non-necrotizing reactions_described in the
mouse (see page 88). Blood flow studies based on the two types ol taction
reveal considerable slowing of blood flow in the centre ot the more;mdoratcd
reactions - a phenomenon that could predispose to tissue necrosis (Potts et al
1992).Jtjs possible that this central slowing of blood flow is due to an effect
^S^S^carappTarance of a positive tuberculin test at 48 hours has
diagnosis of leprosy and tuberculosis are discussed in Chapters 6, 7 and 8,
respectively.
Categories of tuberculin reactors
wzx.isss^sR'.uaz^^*
macrophage cells, is greater in reactions to tuberculin than to
in
patients with tuberculosis, and vice versa in leprosy patients suggesting tl <
this component of the reaction is affected by sPec;es;sPec*fi^Sjthee
diameter of the reaction is not related to the intensity of the cel ular in hl rate
measured as the area of the dermis occupied by the penvascuiar f oci . so
individuals with very large reactions have relatively few cells while others
who have no visible or palpable reaction have an intense cellular infiltrate.
Thus the clinically evident features of the tuberculin reaction are almost cer
tainly due to release of cytokines from the cells rather than to the cell mass
ltSThe practical aspects of skin testing in epidemiological studies and for the
c lients and this may be a genetic feature associated with HLA-D Class 11)
Compatibility antigens (see page 107). There is no mt.den^ that such
non-responders are unable to develop protecttve tmmun ty indeed as men
lioned above, an intense cellular infiltrate may occur in the absence of ch
cally evident reactivity. Category 3 indiv.duals react to some but not al
reagents, indicating that they are responding to the species-specific (Group )
"’^^tapoX^’ints arise from this categorization Hmb an ™a™“sot
the category 2 non-responders is relevant to the use of BCG. Some untortu
nate non-responders have been repeatedly and Linncecssarib' skui tes'ted an
revaccinated in attempts to make them convert. The presence of a BCG scar
sufficient evidence that the vaccmation has ‘taken’. Second, surveys of the
Protective and cross-protective antigens
1
7
♦
vo
reactivity of the category 3 responders is a useful way of determining which
mycobacteria are present in an environment without resorting to the time
consuming procedure of isolating and identifying strains from inanimate
sources. Third, there is an interesting relation between categorization and
disease. Studies on leprosy patients in Nepal (Stanford et al., 1981a) and on
tuberculosis patients in Indonesia (Kardjito et al., 1986) both showed that
most healthy reactive persons are category 1 reactors while the majority of
patients with either disease are category 3 reactors (Fig. 5.9).
In order to understand the relevance of this reduced responsiveness to
common mycobacterial antigens to the development and pathogenesis ot
mycobacterial disease, it is necessary to consider the nature of the protective
epitopes of mycobacteria.
100
80
60
Q
40
20
B
D
E
reactors. Data from Stanford et al. (1981a) and Kardjito et al. (1986)
Protective and cross-protective antigens
It is a widespread belief that ‘protective antigens’, i.e. those that elicit an
effective immune response, are always species- or strain-specific and that vac
cines are ineffective unless they contain such antigens. This is certainly true
when the determinant of virulence is a strain-specific toxin or a viral receptor
site On the other hand, there is no reason to assume that it is necessarily the
case with intracellular pathogens such as the mycobacteria. Indeed the find
ing that BCG is as protective against leprosy as it is against tuberculosis
Vaccination strategies
96
y/
Immunology of mycobacterial disease
(Brown et al., 1966), and the evidence that it protects children against cervi
cal lymphadenitis due to M. avium (Trnka et al., 1994), strongly indicate that
the protective epitopes are to be found among thosecommon to all mycobac^As outlined in Chapter 2, some soluble mycobacterial antigens are actively
secreted by viable bacterial cells. These are probably the first to be presented
to CD4+ T-cells by Class II MHC molecules and could therefore be oi partic
ular relevance to protection involving macrophage activation and granuloma
formation. On the other hand, cylqlytic, activity is.cmcial to protection and
may be induced by a different type of antigen. In this context, a class ol pioteins known as the heat-shock proteins (HSPs) are of particular r^evance.
Heat-shock proteins are present in all living ceils and are structurally highly
conserved. Thus there are many epitopes common to HSPs from mycobac
teria and mammals. Heat-shock proteins are normally present in small
amounts in the cytoplasm and are involved in the assembly and shaping ol
newly synthesized proteins, hence their alternative name chaperonins or
nurse-maid proteins. UndeLConditions of stress, such as heat shock, they^aic
over-produced and expressed on the surface of the cell (Born et al., 1990).
Mycobacterial HSPs may thus be presented on the surfaces of infected cells
but, in addition, intracellular infection, particularly uncontrolled ’Ejection,
may stress the cell leading to over-production and expression ol sell HSPs.
Whatever the origin, HSPs on the cell surface may_actaita^sj^^ytolytic
cells. Indeed a population of CD4+ cytolytic cells recognizing the mycobac
terial 65 kDa HSP have been described (Ottenhoff et al., 1988). In addition, a
major subpopulation of y-8 cells recognize HSPs on cell surfaces and may
therefore destroy stressed, infected macrophages and other cells expressing
HSPs on their surfaces (Orme et al., 1993). As mentioned above (see page
85) these cells may also be required for activation of CD4+ T-cells. It is there
fore likely that mycobacterial HSPs play a central role in protective immunity
It is in this context that the reduced recognition of common mycobacterial
antigens in patients with mycobacterial disease could be relevant as HSPs aic
■ major components of this group of shared antigens.
There is evidence that diminished dermal reactivity to common mycobac
terial antigens is not just a feature of mycobacterial disease but occurs in other
conditions characterized by intracellular parasitism, such as human immuno
deficiency virus (HIV) infection (Khoo et al., 1993) and South American try
panosomiasis (Chagas’ disease; Bottasso et al., 1994). Thus immunotherapy
designed to induce or restore immune recognition of widely distributed anti
gens, which probably are, or include, HSPs might be effective in a range o
unrelated conditions, both infectious and neoplastic, in which recognition and
lysis of abnormal, stressed cells is central to protection (Grange et al., I J Jd).
Vaccination strategies
Many attempts were made to prepare a vaccine against tuberculosis in the
years immediately following Koch’s discovery of the tubercle bacillus As a
result of work by KouLand Trudeau it became a generally accepted dogma
that an effective vaccine would be a live attenuated onejather than a killed
one. Calmette and Gu6rin eventually produced such a
isolataTFSm a case ofTovine mastitis, and therefore presumed lobe M bovts
by passaging it 230 times over a period of 11 years, by which time ext^ive
animafstudies showed it to be stably attenuated These workers P^pared their
viEcine from a bovine rather than a human isolate on account of a behef that
self-limiting tuberculous lesions due to M. bovis in childhood afforded pro
tection against pulmonary tuberculosis later in life (Martan s Law). This vac
cine, Bacille Calmette-Guerin (BCG), was originally given oraHy necmates
able controversy. The results of a number of major vacc,"enl1"a’SJJ^ebfb^
show that protection varies from 80 per cent to none at al . The probable
reason for Ss variation is discussed above (see page 88). When given to an
uninfected (tuberculin negative) child, BCG confers protection agains
serious consequences of subsequent primary infection, such as tuberculous
meningitis. On the other hand, it is muchJess^ffective in preventing tuberc e
bacilli8from persisting in the tissues and causing post-primary tuberculosis
later in life. Thus theyaccine should not be given tojnfected persons, ind _
it is liable to cause severe reactions if given to tuberculin reactors.
■ In view of the inadequacies of BCG vaccine, there is much interest in using
DNA recombinant technology to develop new vaccines. One sugg^e
anproach is t(fidentify the determinants of virulence in M. tuberculosis and to
defete them by, for example, transposon mutagenesis, but this approach would
probably achieve no more than reinventing BCG. Another approach is to-iden
tify protective antigens and to introduce multiple copies of these into BCG or
an alternative vector such as vaccinia.
.
A quite different approach to vaccination is to use an environmental
mycobacterium that never causes disease presumably because it is on y
capable of eliciting protective responses. In this context, BCG, although altenu
atefl, is still capable of causing severe local reactions if given to tuberculinTable 5.2 Results ol nine major BCG vaccine trials
Region
Year of
commencement
Age
range
Protection
afforded (%)
North America'
Chicago, USA
Georgia, USA
Illinois, USA
Puerto Rico
Georgia, USA
Great Britain
South India
South India
1935
1937
1947
1948
1949
1950
1950
1950
1968
0-20 years
3 months
6-17 years
Young adults
1-18 years
5 years
14-15 years
All ages
All ages
80
75
0
0
31
■ Amerindian population
■TlateZ
some protection in those vaccinated in infancy
14
78
31
0”
I
98
Immunological spectra in mycobacterial disease
Immunology of mycobacterial disease
positive persons. Mycobacterium vaccae, a rapidly growing non-pathogen, can
safely be given to such persons and there is accumulating evidence that it is an
effective immunotherapeutic agent for active tuberculosis (sec Chapter 10). Il
able to induce the immune-mediated destruction of persisting mycobacteria in
overt disease it should, by analogy, do so before reactivation occurs and it is
thus a candidate vaccine (Stanford and Grange, 1993).
Attempts have also been made to prepare vaccines against leprosy but, as
the causative organism cannot be cultivated in vitro, it is not possible to pre
pare a living attenuated vaccine from this species. Alternative approaches
include adding killed armadillo-derived M. leprae to BCG and using environ
mental mycobacteria as vaccines. At present, BCG itself is used as a vaccine
against leprosy and high levels of protection have been observed in regions
where it also protects against tuberculosis.
Immunological spectra in mycobacterial disease
The clinical features and course of a mycobacterial disease are, in a very large
measure, dependent upon the immunological reactivity of the patient.
Consequently, the concept of a ‘spectrum’ of such reactivity, from highly
active al one pole to absent at the other, has been developed. At first view this
appears a reasonable and attractive idea but, as in most areas ol mycobacterial
immunology, the matter is not as .straightforward asjt first seems to be.
4a.k
The immune spectrum in leprosy
The great variation in the appearance and behaviour of the determinate forms
of leprosy results from their position on an immunopathological spectrum
described in detail by Ridley and Jopling (1966). For convenience, five points
on the spectrum are recognized: the two polar forms, tuberculoid (TT) and
lepromatous (LL), and three intermediate points, borderline tuberculoid (BT),
mid-borderline (BB) and borderline lepromatous (BL). The clinical, immuno
logical and histological features of these forms are shown in Table 5.3 and arc
described in more detail in Chapter 7.
Table 5.3 Characteristics of the five points in the immunological spectrum of leprosy
Characteristic
TT
Bacilli in lesions
Bacilli in nasal discharge
Granuloma formation
In vitro correlates of CMI
Reaction to lepromin
Anti M. leprae antibodies
Macrophage maturity
Response to therapy
Point on the spectrum
BT
BB
BL
±
LL
+
+
+++
±
++
±
mature
good
immature
► poor
99
At first view, the tuberculoid pole appears to be characterized by effective
immunity which then decreases across the spectrum to the lepromatous pole
where there is no apparent protective immune reactivity. Thus the lesions of
tuberculoid leprosy are characterized by very few bacilli, many lymphocytes
and granulomas containing mature epithelioid cells. In contrast, lesions of
lepromatous leprosy contain few lymphocytes but numerous bacilli within
immature macrophages. Thus, patients with tuberculoid leprosy are some
times regarded as being near normal and those with lepromatous disease as
being the most abnormal. For this reason, shifts in the position of the disease
towards the tuberculoid and lepromatous poles of the spectrum are, respec
tively termed ‘upgrading’ and ‘downgrading’. An alternative view is thQlI
forms of determinate leprosy are equally abnormal, but that the nature of the
abnormality differs^Although tuberculoid leprosy is often self-limiting, the
aranuloma-forming immune response appears to be greatly out of proportion
To the amount of antigen present. It may, in a large part, be a hyper-reactive
response to bacterial debris that is not readily cleared from the lesion. The
mechanism of effective immunity to leprosy must be sought in those contacts
who either never develop the disease or display self-limiting indeterminate
lesions.
,
The factors that determine whether an infected person will develop leprosy
and if so in what form are poorly understood. It was originally considered
that’the outcome of infection is related to the time taken for CMI to develop
(Godal et al., 1974). Thus, in healthy contacts, a rapid onset of CMI would
eliminate the bacilli before lesions had a chance to develop. In tuberculoid
leprosy a slight delay would permit enough multiplication of the bacilli to
render their removal more difficult, while an indefinite delay in the onset of
CMI would result in lepromatous leprosy. The alternative, and more widely
accepted, view is that the type of disease is.‘pre_destined’ by yanous factors,
including genotypes and prior exposure to environmental mycobacteria (see
page 88), and is the result of balances between thewarious T-cell subsets.that
modulate and control immune reactivity.
One of the most extraordinary features of leprosy is the total lack of
immune responsiveness to M. leprae in the lepromatous form of the disease.
This lack of responsiveness is highly specific as patients can respond, some
times very strongly, in skin tests to tuberculin and other mycobacterial
reagents (Stanford, 1994). The defect appears to be due to a suppression of the
activity of M. leprae-speciitc CD4+ T-cells rather than to an absence of these
cells (Bloom et al., 1992). In vitro lymphocyte transformation tests show that
T-helper cells from LL patients respond well to purified antigens of M. leprae
but not to whole bacilli, suggesting that the latter are able, directly or indi
rectly, to suppress T-helper cell activation.
The immunological spectrum in leprosy is related to the Type 1 and lype 2
maturation pathways of T-cells. Thus in TT leprosy the M. /^rae-specific Tcells in the blood are mostly of the TH 1 type and TH 1-associated cytokines
arc detectable in the lesions. In LL leprosy the cytokine response showsa
mixed Type 1 and Type 2 pattern (Yamamura et al., 1991). Thus, in bothlT
leprosy and tuberculosis, a TH 1-mediated response is associated with protec
tive immunity but the effects of the mixed TH1/TH2 responses differ in the
two diseases. In tuberculosis, this mixed response leads to tissue-necrotizing
I
100
8
Immunology of mycobacterial disease
®@®® @y
I
\®®®
@®®b
®
(4)/^
Immunological spectra in mycobacterial disease
Table 5.4 Skin test reactivity to leprosin and tuberculin in healthy individuals and those
with tuberculoid (TT) and lepromatous (LL) leprosy and with tuberculosis all of whom have
been sensitized to both M. leprae and M. tuberculosis
®
® ®
®
®
®
®
Tuberculoid
Group
Reagent
Antigen
groups in
reagent"
Antigen groups
to which
reactivity is
suppressed
Reaction
Health
Leprosin
Tuberculin
Leprosin
Tuberculin
Leprosin
Tuberculin
Leprosin
Tuberculin
i.iv(L)
i.ii.iv(T)
i.iv(L)
i.ii.iv(T)
i.iv(L)
i.ii.iv(T)
i.iv(L)
i.ii.iv(T)
None
None
i
Positive
Positive
Positive
Positive
Negative
Positive
Positive
Positive
TT leprosy
Lepromatous
LL leprosy
Fig. 5.10 The distribution of CD4+ (putative helper phenotype) and CD8+ (putative sup
pressor phenotype) T-cells in lesions of tuberculoid and lepromatous leprosy. Data from
Modlin et al. (1983)
hypersensitivity but this does not occur regularly in leprosy. This appears to
be the result of a strong and specific suppression of immune reactivity: clones
of CD8+ T-cells from lepromatous lesions are able to suppress proliferation
of CD4+ T-cells (putative helper phenotype) specific for M. leprae (Bloom,
1986). Immunocytological studies on the lesions of tuberculoid and leproma
tous leprosy (Fig. 5.10) have shown that there are many T-cells in tuberculoid
lesions but that CD8+ T-cells (putative suppressor phenotype) arc in a minor
ity and occur on the outside of the granuloma. In lepromatous lesions there
are relatively few T-cells, but a much higher proportion of these are of the
CD8+ type and are found within rather than around the lesion (Modlin et al.,
1983).
The ability of LL patients to react strongly to tuberculin while not respond
ing to M. leprae which shares many epitopes with M. tuberculosis suggests
that there are two types of immune defect in these patients. First, there is a
failure to respond to common (Group i) mycobacterial antigens, as is seen in
tuberculosis and some other diseases characterized by intracellular parasitism
(see page 95). Secondly, there is a defect in response to the species-specific
(Group IV) epitopes of M. leprae but not to such epitopes in other mycobac
teria which may thus elicit responses in skin testing and other tests for cellmediated immune reactions. (As noted in Chapter 2, M. leprae is one of a
small group of mycobacteria that possess only Group i and iv antigens: they
lack the Group ii and iii antigens that characterize the slow and rapid growers
respectively.) The skin-test reactivity in patients with tuberculosis and with
tuberculoid and lepromatous leprosy is summarized in Table 5.4.
Further light on the immune defect in leprosy is shed by studies on the socalled leprosy reactions, of which there are two types described in detail in
Chapter 7. Type 1 reactions occur in patients with borderline forms of leprosy
(BT, BB, BL) and are often associated with a shift in immunity towards the
TT pole. The cytokines detected in Type 1 reactions are predominantly TH1induced, suggesting that the reactions are associated with a TH2 to TH1 shift
(Stanford, 1994). Type 2 reactions, erythema nodosum leprosum (ENL),
nrrnr in natientc at nr nnar thr> T T r-.nl.-> Th« nxtnl’innc
101
Ucinnc nm nf n
Tuberculosis
i.iv(L)
* i = common to all mycobacteria; ii = slow grower specific (absent from M. leprae}-, iv(L)
antigens specific to M. leprae: iv(T) antigens specific to M. tuberculosis (see Fig. 2.1,
page 13)
mixed Type 1 and Type 2 pattern, rellecling the underlying leprosy-specific Tcell maturation pattern at this end of the spectrum. Though ENL is usually
said to be an example of an antigen-antibody complex reaction, TNFmediated tissue necrosis is involved and is probably the initiating event. This
appears to be due to a relaxation of suppression by CD8+ cells (Filley et al.,
1989).
Although details are far from clear, it appears that LL leprosy patients are
protected against tissue-necrotizing hypersensitivity of the Koch type by the
specific CD8+ T-cell-mediated suppression described above. On the other
hand, they lack the pure TH1 cytokine response that would lead to granuloma
formation, bacillary destruction and ultimate resolution of the disease. In this
respect, injection of microgram amounts of gamma interferon (a TH1
cytokine) into the skin of LL patients leads to an increase in the number of
CD4+ cells relative to CD8+ cells, granuloma formation and destruction of
leprosy bacilli, indicating a shift towards theTT pole of the spectrum (Samuel
etal., 1987).
For more details on the spectrum and immunopathology of leprosy see
Ridley (1988).
The immune spectrum in tuberculosis
In view of the description of the immune spectrum in leprosy, the existence of
a similar spectrum in tuberculosis has been postulated by several workers.
Ridley and Ridley, for example, divided 54 patients with tuberculosis into
three histological groups, each with two subgroups (Table 5.5; Ridley and
Ridley, 1987). In common with leprosy, immune reactivity, including the
maturity of the macrophages in the lesions, and the bacillary load are
.-cr>I, > rnlntort
Immunological spectra in mycobacterial disease
102
Immunology of mycobacterial disease
The immune spectrum in Mycobacterium ulcerans infection
Table 5.5 The spectrum of tubercuto^-OalaJrornJjdley and Ridley (19871
Group Principal
cell type
la
1b
2a
2b
3a
3b
Organized mature
epithelioid cells
Unorganized mature
and immature
epithelioid cells
Immature epithelioid
cells
Immature epithelioid
cells and/or
undifferentiated
histiocytes
Scanty macrophages
Very few
macrophages
Necrosis
None
Patchy fibrinoid
Caseation, no
nuclear debris
Necrosis with
nuclear debris and
polymorphs
Extensive, basophilic;
coarse nuclear debris
Extensive,
eosinophilic; scanty
nuclear debris
Giant
cells
Bacilli
None
Rare
Scanty
+++
spectra of tuberculosis and lep V (
specific defect seen in leproassociated with iramu"°^®,Ca j (
anergy intaberculosis is usually more
matous leprosy.does nW occur Ins tead
mv infection (see
Akhouah CD8+ cdls are found in lesions of severe, progress.ve
“emulosis^(Ailsl.® «/., 1992), it is uncertain whether they are a cause or a
Immunological reactivity in this relatively uncommon disease is
and
fascinating The course of the disease is shown in F g. 5.11. The lesion,
more fully described in Chapter 9, commences as a skin nodule that may
e™her resolve or progress to overt ulceration. Unlike other pathogenic
mvcobacteril a major determinant of the virulence of M. ulcerans is a toxin
which in the progressive cases, causes widespread necrosis and liquefaction
of the subcutaneous fat (Hockmeyer et al., 1978) Secondary necrosis of the
overlving skin results in deeply undermined ulcers, often reaching enor
mous sizes. During the progressive ulcerative stagethere is evidence of
immunological anergy as the lesions contam many bacdli bu th^ '-me
or no cellular response and the patient fails to react to Burn! n, a skin test
reagent prepared from the causative organism (Stanford er <1975TThe
inersv is less specific than in lepromatous leprosy as patients fad to react to
tuberculin as well as to Burulin. There is some evidence that this anergy is
due to trapping of T-cells responding to mycobacterial antigen in the lymp
“unless the lesion is excised, a stage is reached when an effective immune
resnonsc ensues lymphocytic infiltrates and granuloma formation are seen,
the bacilli decline in number and then disappear, the patient reacts to Burulin
and tubercuhn and the lesion eventually heals by h -is^e^use of the
characteristic shift from anergy to immune reactivity in this disease
understood It appears to be a local phenomenon as a lesion may show^neJgy
in one part and healing in another. Il would be of interest to determine whether
this shift is due to changes in local cytokine patterns related to TH. and
T-cells.
Scarring
Ulcerative-----------------
Pre-ulcerative
4- Progressing-------- -------- Resolving------ ►
Infection
^S"on,^‘its’relationio erythema nodosum leprosum, is not
i
_ 4
103
Immune shift
shaded area indicates the extent of clinically evident lesion
♦
I
104
Mycobacterial adjuvants
Immunology of mycobacterial disease
Immunotherapy
Many attempts have been made to treat tuberculosis and leprosy by stimulat
ing the patient’s immune reactivity. Koch’s use of Old Tuberculin was the fust
such attempt and it appeared to be of value in the case of skin tuberculosis.
Many other attempts followed, with some claims of success.
Until recently, attempts at immunotherapy for mycobacterial disease weic
mostly based on non-specific stimulation of immune responsiveness b_y_means
of myeixbacie.rUlUmigensor mimunostimulaiing drugs such as levamisole. As
it is now clear that there arc fundamental differences between protective anc
tissue-necrotizing immune responses, it is important that immunotherapy
should enhance the former and, hopefully, suppress the latter. Following
observations that BCG is very effective as a vaccine against leprosy in parts
of Uganda and that the efficacy of this vaccine is affected by prior sensitiza
tion of the population by environmental mycobacteria, a number of these
species were isolated from Ugandan mud and extensively studied. One
species M. vaccae, was found to down-regulate the Koch phenomenon and to
restore skin test reactivity to the common mycobacterial antigens (Stanford et
al. 1994). Although originally intended as a leprosy vaccine, it was ‘^una to
be’a useful agent for the immunotherapy of tuberculosis as described in
Chapter 10. This immunotherapeutic agent is a TH 1 adjuvant and thus induces
protective cell-mediated immune reactions and, by restoring recognition o
common antigens including heat-shock proteins, it probably also Iacihtales
recognition and killing of stressed, mycobacteria-laden cells. It is likely that
all these mechanisms are secondary to shifts in the steroid hormone balance in
the lymphoid microenvironment (Rook et al., 1994; see page 89).
I
t..
Humoral factors and serodiagnosis
It is unlikely that antibody plays a significant part in the immune response to
mycobacteria. Children with disseminated tuberculosis have lower levels ol
antibody to mycobacterial lipoarabinomannan (LAM) than those with local
ized disease but the relation between this antibody response and pathogenesis
is not clear (Costello et al., 1992). Other correlations between antibody levels
and disease status may merely reflect the antigen load or the maturation path
way of the mycobacteria-specific T-cells as TH2 cells facilitate antibody pro
duction but TH 1 cells do not.
A serological characteristic of active tuberculosis and some autoimmune
diseases characterized by tissue necrosis is an^v^d jevel of agalactosyl
immunoglobulin G (Gal(O); Rook, 1988). This is imn^npglobuhn G (IgG)
which lacks the terminal galactose on the sugar chain in the CH,, domain ot me
heavy chain. It accounts for 20-25 per cent of the total circulating IgG, rising
to 40 per cent or more in the elderly. Its function is unknown but it appears to
be an indicator of Koch-type tissue necrotizing reactions and its decline is a
useful marker of the efficacy of immunotherapy for tuberculosis (Rook et al.,
1994).
Though of limited relevance to pathogenesis, there have been numcious
® ....
i
i •
......... r....
l.nimoei ir
105
(Grange, 1984; Wilkins, 1994). Despite a huge amount of effort, no serodiagnostic test is in widespread clinical use. The reason for this is that, although
antibodies are undoubtedly produced in mycobacterial disease, the overlap
between levels in patients and either healthy individuals or those with other
diseases is unacceptably large. ‘Natural’ antibodies are probably due to con
tact with mycobacteria and related genera in the environment, but the use of
purified specific antigens in diagnostic tests has proved disappointing, most of
the humoral immune response is directed towards shared antigens. The intro
duction of monoclonal antibodies made it possible to develop competition
assays for antibody to specific epitopes. Although this approach led to high
specificity (i.e. very few ‘false positives’), the sensitivity (the ability to diag
nose disease when present) was rather low (Wilkins, 1994), although the use
of a set of several monoclonal antibodies, while making the test rather com
plex, improved the sensitivity (Hoeppner et al., 1987).
Serodiagnosis of leprosy has been attempted by using the PGL-1 antigen
but, as is explained in Chapter 7, only patients with multibacillary disease
have antibody levels significantly higher than those of healthy contacts.
The use of antibody to detect mycobacterial antigen, and antigen-antibody
complexes, in clinical specimens has not received as much attention as serodiagnosis although the results of a limited number of studies are encouraging,
with high sensitivity and specificity. In general, the best results have been
obtained with ‘clean’ specimens such as cerebrospinal, pleural and peritoneal
fluids (Wadee et al., 1990).
Mycobacterial adjuvants
An adjuvant is a substance that enhances the antibody response to unrelated
antigens administered with it. One of the best known is Freund s complete
adjuvant which consists of a water-in-oil emulsion containing killed tubercle
bacilli The adjuvant activity of mycobacteria resides principally in the pepti
doglycan (murein) of the cell wall, although trehalose dimycolate (cord fac
tor) RNA and certain peptidoglycolipids also appear to possess this activity.
In the case of peptidoglycan, the minimum structure required for adjuvant
activity is muramyl dipeptide (Chapter 2, page 17), a water soluble molecule
that has been synthesized.
.
Despite many studies, the mode of action and natural function of adjuvants
remain shrouded in mystery. Reactions to them may be associated with a
primitive immune recognition system which responds to certain common
microbial components. In addition to enhancing antibody production, adju
vants affect T-cells and can thereby cause some degree of macrophage activa
tion Indeed this property might be of relevance to the containment of
mycobacterial infections before the onset of specific cell-mediated immunity
as muramyl dipeptide induces granulomas very similar in appearance to those
of immune origin in tuberculosis. Theories as to the mode of action of adju
vants include assistance in antigen presentation to T-cells, effects on lympho
cyte traffic and non-specific mitogenic effects on B-cells. They may also act
by modulating the steroid hormone balance in the lymphoid microenviron
ment (see naue 89).
I
Mycobacterial disease and immunosuppression
106
107
Immunology of mycobacterial disease
Mycobacterial disease and autoimmunity
There has been considerable interest in a possible relationship between infec
tion by mycobacteria and autoimmune disease. Freund’s complete adjuvant (a
suspension of killed mycobacteria in oil: see above) induces autoimmune
arthritis in rats. Some patients with tuberculosis develop a s_tenle_arthntis
(Poncet’s disease; see page 174), although this is uncommon.
There are some other fascinating links between mycobacterial and auto
immune diseases (Rook and Stanford, 1992). Thus in common with tubercu
losis, patients with rheumatoid arthritis (RA) have elevated levels of
aealactosyl IgG (Rook, 1988). Also, patients with RA are often of the HLADR4 phenotype which is associated with strong dermal reactivity to species
specific antigens of M. tuberculosis (see below, page 107).
As mycobacterial heat-shock proteins (HSPs) share many epitopes with
their human analogues they could, in theory, induce autoimmune phenomena.
(Das and Grange, 1993). Indeed the arthritis inducible in rats by Freund s
complete adjuvant is adoptively transferable by T-cells reacting to the
mycobacterial 65 kDa HSP (van Eden et al., 1988). In practice, there is no
firm evidence that HSPs or other mycobacterial components induce au oimmune phenomena that contribute significantly to immunopathology and it
has been postulated that they arc usually prevented from doing so by an elab
orate regulatory system (Cohen and Young, 1991).
Genetic factors in mycobacterial immunity
As only a minority of those infected with mycobacteria develop overt disease,
many attempts have been made to establish some genetic marker of suscepti
bility or resistance. Studies on identical and non-identical twins stiongly sug
gest an inherited predisposition or resistance to tuberculosis and racial
variations have been suggested. Extensive studies in the mouse have revealed
that an allele, designated Beg, confers natural macrophage-medmted resis
tance to a range of intracellular pathogens including BCG, salmonellae and
Leishmania donovani (Skamene et al., 1982) This allele codes for a protein
termed natural-resistance-associated macrophage protein (flramp} which s
involved in the generation of reactive nitrogen intermediates (Vidal et al.,
J993). A human homologue of this protein has been found but its significance
to protection is unknown. As doubt has been shed on the ability of human
macrophages to generate sufficient levels of reactive nitrogen intermediates to
kill mycobacteria (see page 82), this gene may not play such a key role in
resistance to disease in man as it does in mice.
.
,
Many attempts have been made to link susceptibility of tuberculosis and
leprosy to the Class I antigens (HLA-A and HLA-B) of the major histocom
patibility complex. Although some studies show a low but significant associ
ation of a particular HLA type to overt disease, the results vary from ’egton to
region and no definite pattern has emerged. In the case of Class II (HLA-D)
antigens, it has been shown that HLA-DR2, particularly the DR15 subtyP^
predisposes to the development^ of tuberculosis,jarWulartyradto ogt^^y-
et al., 1991; Khomenko et al., 1990). The HLA-DR2 specificity may affect
antigen recognition as individuals of this genotype have higher levels of anti
body to epitopes on a 38 kDa protein unique to M. tuberculosis than those
lacking this genotype (Bothamley et al., 1989).
In leprosy, HLA-DR3 predisposes to tuberculoid leprosy and protects
against lepromatous leprosy while HLA-DQwl predisposes to the develop
ment of lepromatous leprosy (Ottenhoff and de Vries, 1987).
Class II types also affect skin test responses to mycobacterial antigens.
Sensitized individuals who, nevertheless, do not respond to testing with such
antigens (category 2 non-responders; see page 93) do not express the HLADR3 specificity (van Eden et al., 1983) while those of HLA-DR4 phenotype
respond relatively strongly to species-specific antigens of M. tuberculosis but
not to such antigens of other mycobacterial species (Ottenhoff et al., 1986).
Larger tuberculin reactions were also found in HLA-DR 15-positive than in
DR 15-negative tuberculosis patients (Bothamley et al., 1995).
Mycobacterial disease and immunosuppression
Immunosuppression of whatever cause predisposes to the development of
tubciiulQSjs, whether due to primary infection or to endogenous reactivation
or exogenous reinfection. Since the early 1980s, HIV has become an
increasingly common predisposing factor in tuberculosis as described in
detail in Chapter 8. In addition, HIV infection predisposes to disease due to
environmental mycobacteria, notably the M. avium complex (see below and
Chapter 9).
,
.
There is no doubt that HIV infection has a profound effect on tuberculosis
and there is evidence that the latter disease adversely affects the course of the
former. It has been observed that, evetLitsuccessfully treated, tuberculosis in
an HIV-positive person has a very deleterious effect on_fu.Lure health (see
Chapter 6). Indeed, tuberculosis appears to drive the patient into the full pic
ture of AIDS with considerable shortening of life. While the details are not
clear, there is evidence that TNF-a and other immunological mediators
released in tuberculosis lead to transactivation of the HIV provirus and its
subsequent replication (Osborn et al., 1989). In addition, tuberculosis causes
a CD4+ T-cell lymphopenia which may add to that induced by the HIV (Beck
eTaT, 1985). Whatever the cause, the occurrence of active tuberculosis in the
HIV-’posilive patient has very serious consequences, demanding strenuous
efforts to prevent such disease by programmes of chemoprophylaxis or,
■preferably, immunoprophylaxis.
Not only is there a reduced ability to develop the characteristic high
turnover granuloma of immunogenic origin in immunosuppressed tubercu
losis patients, there is also a suppression of tissue-necrotizing reactions and
scar formation that would otherwise limit the spread of infection. Thus, discrete“pulmonary lesions and cavity formation are both less common in such
patients. Instead, there may be radiologically rather non-specific spreading
pulmonary lesions (Fig. 5.12). Non-pulmonary lesions due to unrestricted
bacillary dissemination are frequent in such patients, particularly in the more
profoundly immunosuppressed, and may present as one or more solitary
Summary - an integrated view of mycobacterial immunity
108
109
Immunology of mycobacterial disease
In contrast to tuberculosis, evidence for an interaction between HIV and lep
rosy is limited and conflicting. In general, clinical leprosy is no more ^equenl
in HIV-positive than in HIV-negative individuals and, among leprosy patients
there is no relation between HIV status and the bacterial load It has been
observed that certain infections, notably those due to parasites and woi ms, are
not adversely affected by HIV infection (Lucas, 1990), and it has, been^postu
lated that leprosy may be among these ‘missing infections (Lucas, 1993).
Summary - an integrated view of mycobacterial
immunity
The host’s mechanism for defence against mycobacterial disease resembles an
orchestra with some lead performers such as the antigen-specific lymphocytes
and macrophages and others with minor roles. In the past, the PnnclPa
defence mechanism was thought to be the activated macrophage P«r
attempts to demonstrate effective killing of M. lube^‘S
,h^f^
macrophages have been elusive. It appears much more likely that effective
defence'requiresJhe aggregation of many macrophages, thereby ‘“'""ng
granuloma Although individual macrophages within the granuloma may well
mhibit bacterial growth and kill some, the very hostile anoxic and acidic
environment in the centre of the lesion may be much mgre important m over
coming the infection. In this respect, there are several immune mechanisms
for seeking mycobacteria-laden macrophages and other cells and lysing them.
These include mUupaLkiller cells, Yn8
CDStfand some CD4+) cells. The Largets for such lysis may be spec he
mycobacterial, antigen oLendogenous heat-shock proteins presen ted on the
snftoof stressetLcells. In addition, healing processes surround the lesions
...i:li dense >ca: tissue, thereby enhancing the hostility
ment to any residuaLmycobactena and eventually walhng off the resolving
^chlmmune mechanisms are usually very effective ^ingW^olution
to M. tuberculosis. Courtesy of Dr P. Ormerod
lesions, as widespread lymphadenopathy or as multi-organ involvement. The
latter differs from miliary tuberculosis as the discrete granulomas do no
develop. Instead, organs contain minute necrotic foci teeming with acid-fast
bacilli These may not be visible radiologically and are only detected by
biopsy or at autopsy. Thus, this form of the disease has been termed cryptic
disseminated tuberculosis’ (Proudfoot, 1971).
of the initial lesions of tuberculosis and leprosy in about 95 per cent ol^cases.
Unfortunately they seem unable Lo pie^ent the mycobacteria trom entenn
the poorly understood ‘persistor’ slate which may be responsible for post
primary disease in a minority of infected individuals years or decades later.
The protective immune responses are initiated by cytokines
THl T-cells, notably IFN-y although
mation and subsequent fibrosis. Post-primary tuberculosis, which occurait
about.5 percent of infected individuals, is associated with a mjxed
T-cell response which has the effect of rendering cells exquisitely s^ceptiblc
to killing by TNF-a. This results in excessive tissue-necrotizing hypersensi
livity which generates the cavities, without which transmission of the disease
would not occur. Another feature of active tuberculosis is the suppressed
immune recognition of common mycobacterial epitopes mdudm those on
heat-shock proteins. The exact significance of this ‘immunological blindness
to protection is not clear but it could inhibit or delay the recognition and lysi
of stressed cells laden with replicating bacilli.
!
110
Immunology of mycobacterial disease
The spectrum in leprosy is likewise related, at least in part, to the type of Tcell, TH1 or TH2. In tuberculoid leprosy the TH1 response causes formation
of granulomas in nerves, thereby damaging them, and much ot this reactivity
may be aimed at bacterial debris that is not easily cleared form the tissues, in
lepromatous leprosy the mixed TH1/TH2 response does not give rise to
massive tissue necrosis as it does in tuberculosis because of a high degree oi
antigen-specific immune suppression mediated by CD8+ T-cells. A break
down of this suppression appears to be responsible for leprosy reactions.
Many other factors are involved in immune responses to mycobacteria: the
interactions of cells and cytokines are exceedingly complex. Nevertheless,
there is increasing evidence that the ‘choice’ between health and progression
to post-primary tuberculosis or multibacillary leprosy is related to the matura
tion pathway of the T-cells. This choice may be affected by genetic and, prob
ably more critically, by environmental factors such as exposure to
mycobacterial antigens that imprint the immune system with a pattern of sub
sequent responsiveness. There is evidence that this c^mpLejumniunQlp^ical
orchestra is conducted by the endocrine system which determines the balance
bet^STh^So^s promotingJTHLand TH2 Tcel.l_matiLration pathways.
There may indeed be a very complex interaction between antigen recognition,
adjuvant activity, hormonal function and regulatory centres, such as the hypo
thalamus, in the brain. Whatever the nature and complexity of the regulatory
system, the logical approach to the prevention and treatment of mycobacterial
disease is to attempt to switch immune responsiveness from a mode that
causes progressive disease to one that facilitates protection.
I.
References
References
111
world (American Association of Immunologists Presidential Address). Journal of
Immunology 137, i-x.
.
Bloom, B.R. (ed.) 1994: Tuberculosis: pathogenesis, protection, and control.
Washington DC: American Society for Microbiology.
Bloom, B.R., Salgame, P. and Diamond, B. 1992: Revisiting and revising suppressor
T cells. Immunology Today 13, 131-6.
Boom WH, Wallis, R.S. and Chervenak, K.A. 1991: Human Mycobacterium
tuberculosis-veactive CD4+ T-cell clones: heterogeneity in antigen recognition
cytokine production, and cytotoxicity for mononuclear phagocytes. Infection and
Immunity 59, 2737-43.
Born, W„ Hall, L., Dallas, A. et al. 1990: Recognition of a peptide antigen by heat
shock reactive y-8 T lymphocytes. Science 249, 67-9.
Bothamley, G.H. and Grange, J.M. 1991: The Koch phenomenon and delayed hyper
sensitivity 1891-1991. Tubercle 72, 7-12.
Bothamley, G.H., Beck, J.S., Geziena, M. et al. 1989: Association of tuberculosis and
M. tuberculosis-specific antibody levels with HLA. Journal of Infectious Diseases
159 549-55
Bothamley, G.H., Gibbs, J.H., Beck, J.S. et al. 1995: Delayed hypersensitivity and
HLA in smear-positive pulmonary tuberculosis. International Archives of Allergy
and Immunology 106, 38^45.
.
Bottasso, O.A., Ingledew, N., Keni, M. et al. 1994: The cellular immune response to
common mycobacterial antigens is suppressed in persons seropositive tor
Trypanosoma cruzi- Lancet 344, 1340-1.
e i
Brahmajothi, V., Pitchappan, R.M., Kakkanaiah, V.N. et al. 1991: Association of pul
monary tuberculosis and HLA in South India. Tubercle 72, 123-32.
Bretscher, P.A. 1992: A strategy to improve the efficacy of vaccination against
tuberculosis and leprosy. Immunology Today 13, 342-5.
Brown, J.A.K., Slone, M.M. and Sutherland, I. 1966: BCG vaccination of children
against leprosy in Uganda. British Medical Journal 1, 7-14.
Chan, J., Fan, X., Hunter, S.W. et al. 1991; Lipoarabinomannan, a possible virulence
factor involved in persistence of Mycobacterium tuberculosis within macrophages.
ChrrK^/rM^^Rl' and Bloom, B.R. 1992: Killing of virulent
forms of human pulmonary tuberculosis at presentation and during recovery.
Al Al^^Nforino. C. and Rook, G.A.W. 1992: TNF-alpha-medialed tissue
damage in mouse footpads primed with mycobacterial preparations. Research m
BaSt^cT, Schreiber, R.D. and Unanue, E.R. 1991: Natural
independent pathway of macrophage activation defined in the SCID mouse.
B “Xe. al. .986: The cellular responses of tubercu
losis and leprosy patients and of healthy controls in skin tests to new tuberculin and
leprosin A. Clinical and Experimental Immunology 64,484-9.
Beck, J.S., Morley, S.M., Lowe, J.G. et al. 1988: D.versity in migration of CD4 and
CDS lymphocytes in different microanatomical compartments of the skm i
c
tuberculin reaction in man. British Journal of Experimental Pathology 69, 771-80
Beck J.S., Potts, R.C., Kardjito, T. and Grange, J.M. 1985: T4 lymphopenia in
patients with active pulmonary tuberculosis. Clinical and Experimental
BeX b'°G “nw^D., Hulmes. W. e, al. 1985: Identity of tumour necrosis
factor and the macrophage-secreted factor cachectin. Nature 316, 552 3.
Bloom, B.R. 1986: Learning from leprosy: a perspective on immunology and the thud
Mycobacterium tuberculosis by reactive nitrogen intermediates produced by acti
vated murine macrophages. Journal of Experimental Medicine 175, 1111 22.
Cohen. I.R. and Young, D.B. 1991: Autoimmunity, microbial immunity and the
immunological homunculus. Immunology Today 12, 105-10
.
Costello, A.M.L., Kumar, A., Narayan, V. et al. 1992: Does antibody to mycobacterial
antigens, including lipoarabinomannan, limit dissemination m childhood tuberculosis.
Transactions of the Royal Society of Tropical Medicine and Hygiene 86, 686-92.
Dannenberg, A.M. 1993: Immunopathogenesis of pulmonary tuberculosis. Hospital
Practice 28,51-8.
.
.
Das, P.K. and Grange, J.M. 1993: Mycobacteria in relation to tissue immune response
and pathogenesis. Reviews of Medical Microbiology 4, 15-23.
Daynes, R.A , Meikle, A.W. and Araneo, B.A. 1991: Locally active steroid hormones
may facilitate compartmentalization of immunity by regulating the types of
lymphokines produced by helper T cells. Research in
40-5_
Draper. P. 1981: Mycobacterial inhibition of intracellular killing. In O Grady, F and
Smith, H. (eds.), Microbial perturbation of host defences, pp. 143-64. London:
Academic Press.
Editorial. 1984: New tuberculins. Lancet 1, 199-200.
El Ansary, E.H. and Grange, J.M. 1984: Qualitative differences in tuberculin reactivity
in patients with tuberculosis, occupational contacts and non-contacts. Tubercle 65,
191-4.
112
Immunology of mycobacterial disease
Filley E Andreoli, A., Steele, J. etal. 1989: A transient rise in agalactosyl IgG correladnr'wkh free interleukin-2 receptors during episodes of erythema nodosum
\^umXrmicalandExperin^
•
Filipv FA Bull HA., Dowd, P.M. and Rook, G.A.W. \))2.. me eneci
Xtoeriutn \uberculosis on the susceptibility of human cells to the sfmulatory
FinaeTE MeX Im—iXd m mbS'i^ictio. for pathogenesis
F and^nZt In Porter, J.D.H. and McAdam K.RW.J. (eds.), Tuberculosa back
I
counts. IX International Conference on AIDS, Berlin, June 6 11. 1993. Abstract
Koch^ Pic ^891: Weitere Mitteilungen liber ein Heilmittel gegen Tuberkulose.
Deutsche Medizinische’
rm
1976’ Tuberculin sensitivity of
Flesch YrraS Kaufmanm^KE.6!990: Activation of tubercutostatic; ma^P^
F functions by gamma interferon, interleukin-4 and tumour necrosis factor. Infection
Flvnn ^L^GoldsVn/M’M. Triebold, KJ. et al. 1992: Major histocompatibility
comnlex’class I-rcstricted T cells are required tor resistance: to Mycobactenuni
tuberculosis infection. Proceedings of the National Academy of Science of the US
Gibbs1??!13Grange J.M., Beck, J.S. etal. 1991: Early delayed-type hypersensitivity
responses in tuberculin skin tests after heavy occupational exposure to tuberculosis.
Goll^‘b Mj'rval^S^
Samuel, D.R. 1974: Recent advances in
the immunology of leprosy with special reference to new' approaches
approaches in
in immu
prophylaxis. Bulletin de ITnstitut Pasteur 72, 273-310.
Grange, J.M. 1984: The humoral immune response in tuberculosis: its nature.
biological role and diagnostic usefulness. Advances m Tuberculosis Research 21,
1-7!' T M Rprk I 8 Haroer E.I. et al. 1986: The effect of exposure of hospital
'employees’to patients with tuberculosis on dermal reactivity to four new tuber-
L
oh k' A Kress Y and Bloom, B.R. 1993: Pathogenesis of tuberculosis
1 interactions of Mycobacterium tuberculosis with macrophages. Infection an
and Rea TH 1983: T-lymphocyte subsets
Immunity’61, 2763-73.
°f
M^eS KfX-W —
MoTrnZ^T and Xre. K.W. 1991: The role: of IL-10 in erossregulation of Th,
audThlrespo^
of the MycobacIerium
O^mXXn,'R<)and Boom. W.H. 1993. T cell response to Mycobacterium
faclor and inlerleukin
iuberculos^. Jourim‘Infectious^
GrSX^^SUntaK and Rook. G.A.W. 1995. Tuberculosis ami cancer:
parallels in host responses and therapeutic approaches? l^ccrM5^ ^0-2.
Ot^molL TH.M, and de Vries, R. 1987: Recognition of M. leprae antigens.
1988-The recombinant
Dordrecht: Martinus Nijholf.
i-onoW^num
“ami’body-based competition test for the serology of tuberculosis in Indonesia.
«s of active tuberculosis by i.—
S ^^XZZtS^lis.
the pathogenesis of rheumatoid arthnt£
3 10^
mycobacleria on
’’“KG vacctation a8nd tuberculosis. American Review of Respiratory Disease 94.
553-68.
r.ihhs J H. et al. 1992: Measurements of blood How and
Jutacuiosis Ind healthy subjects. Europe.,, Journal of Respiratory Diseases 69,
TZ 14?-5' c u F Rinm C and Yamamoto S. 1993: Crosstalk between oc/p T cells
^Slcens^^^aSivatS^Tiel! responses afteryffiTceh mod^
with7the monoclonal antibody GL3. Proceedings of the National Academy of
Khomenko^tG^’vm^^
Chukanova, V.P. and Pospelov L.E >990:
Tubercu'osB^
of
^c^mmon mycobacterial antigens in HIV-infected individuals despite high CD4
ProudfoU AX 1971: Cryptic disseminated tuberculosis. British Journal of Hospital
oACR\n’d7McCordock HA 1929: An enquiry concerning the role of allergy.
R,XunV"ndl^mh^factors^importance in the2Pat5enesis of human tuberculosis. Bulletin of the Johns Hopkins Hospital 44, 273^22.
References
114
115
Immunology of mycobacterial disease
Stanford, J.L., Stanford, C.A., Rook, G.A.W. and Grange, J.M. 1994: Immunotherapy
for tuberculosis. Investigative and practical aspects. Clinical Immunotherapeutics
1,430-40.
j
Sylvius, F. 1680: Opera medico (Tractatus IV: De phthisi). Geneva: deTournes. 526.
Trnka, L„ Dankova, D. and Svandova, E. 1994: Six years’ experience with the dis
continuation of BCG vaccination. 4. Protective effect of BCG vaccination against
the Mycobacterium avium-intracellulare complex. Tubercle and Lung Disease 75,
348—52
van Eden, W., De Vries, R.R.P., Stanford, J.L. and Rook, G.A.W. 1983: HLA DR3
associated genetic control of response to multiple skin tests with new tuberculins.
Clinical and Experimental Immunology 52, 287^-92.
van Eden, W., Thole, J.E.R., van der Zee, R. et al. 1988: Cloning of the mycobacterial
epitope recognised by T lymphocytes in adjuvant arthritis. Nature 331, 171-3.
Vidal, S.M., Malo, D., Vogan, K. etal. 1993: Natural resistance to infection with intra
cellular parasites: isolation of a candidate for Beg. Cell 73, 469-85.
von Pirquet, C. 1907: Demonstration zur Tuberculindiagnose durch Hautimpfung.
Berliner Klinische Wochenschrift 48, 699.
Wadee, A.A., Boting, L. and Reddy, S.G. 1990: Antigen capture assay for detection of
a 45-kilodalton Mycobacterium tuberculosis antigen. Journal of Clinical
Microbiology 28, 2786-91.
.
Wilkins, E.G.L. 1994: The serodiagnosis of tuberculosis. In Davies, P.D.O. (ed.).
Clinical Tuberculosis. London, Chapman & Hall. 367-80.
Yamamura, M„ Uyemura, K., Deans, R.J. et al. 1991: Defining protective responses
to pathogens: cytokine profiles in leprosy patients. Science 254, 277-9.
Ridley D S 1988: Pathogenesis of leprosy and related diseases. London: Wright.
Ridley, D.S. and Jopling, W.W. 1966: Classification of leprosy according to immunity:
a five group system. International Journal of Leprosy 34, 255—73.
Ridley, D.S. and Ridley, M.J. 1987: Rationale for the histological spectrum of tubercu
losis. A basis for classification. Pathology 19, 186-92.
.
Rook, G.A.W. 1986: The role of vitamin D in tuberculosis. American Review oj
Rook^GjVWH^SS: Rheumatoid arthritis, mycobacterial antigens and agalactosyl
IgG. Scandinavian Journal of Immunology 28, 487-93.
Rook, G.A.W. and Al Attiyah, R. 1991: Cytokines and the Koch phenomenon.
Tubercle 72, 13-20.
. .
,
, - T
Rook, G.A.W. and Bloom, B.R. 1994: Mechanisms of pathogenesis in tuberculosis. In
Bloom, B.R. (ed.), Tuberculosis: pathogenesis, protection, and control. Washington
DC: American Society for Microbiology. 485-501.
Rook, G.A.W. and Stanford, J.L. 1979: The relevance to protection of three forms of
delayed skin test response evoked by M. leprae and other mycobacteria in mice: cor
relation with the classical work in the guinea-pig. Parasite Immunology 1, 111-23.
Rook, G.A.W. and Stanford, J.L. 1992: Slow bacterial infections or autoimmunity.
RoSTg A W^Onyebioh! R^d Stanford, J.L. 1993: THimW switching and loss of
CD4 cells in chronic infections: an immuno-endocnnological hypothesis not exclu
sive to HIV. Immunology Today 14, 568-9.
.
Rook G.A.W., Onyebujoh, R, Wilkins, E. et al. 1994: A longitudinal study of per cent
agalactosyl IgG in tuberculosis patients receiving chemotherapy, with and without
immunotherapy. Immunology 81, 149-54.
.
Rook GAW Steele, J., Ainsworth, M. and Champion, B.R. 1986: Activation of
macrophages to inhibit proliferation of Mycobacterium tuberculosis; companson of
the effects of recombinant gamma interferon on human monocytes and munne pen
toneal macrophages. Immunology 59, 333-8.
Samuel, N.M., Grange, J.M., Samuel, S. et al. 1987: A study of the effects of intra
dermal administration of recombinant gamma interferon in lepromatous leprosy
Seibert^EB^m?TheTsolafion^properties of the purified protein derivative of
tuberculin. American Review of Tuberculosis 30, 713—20.
Skamene, E„ Gros, P., Forget, A. et al. 1982: Genetic regulation of resistance to intra
cellular pathogens. Nature 297, 506-9.
,
Skinsness, O.K. 1968: Comparative pathogenesis of mycobactenoses. Annals oj the
New York Academy of Science 159, 19-31.
Stanford, J.L. 1994: The history and future of vaccination and immunotherapy tor
leprosy. Tropical and Geographical Medicine 46, 93-107.
.
Stanford, J.L. and Grange, J.M. 1993: New concepts for the control of tuberculosis in
the twenty first century. Journal of the Royal College of Physicians 27, 218 23.
Stanford, J.L. and Lerna, E. 1983: The use of a sonicate preparation of Mycobacterium
tuberculosis (new tuberculin) in the assessment of BCG vaccination. Tubercle 64,
275—82
Stanford, J.L., Nye, P.M., Rook, G.A.W. et al. 1981a: A preliminary investigation of
the responsiveness or otherwise of patients and staff of a leprosy hospital to gioups
of shared or species specific antigens of mycobacteria. Leprosy ^view 52, 321-7.
Stanford, J.L., Revill, W.D.L., Gunthorpe, W.J. and Grange, J.M. 1975: The produc
tion and preliminary investigation of Burulin, a new skin test reagent tor
Mycobacterium ulcerans infection. Journal of Hygiene 74, 7-16.
Stanford, J.L., Shield, M.J. and Rook, G.A.W. 1981b: Hypothesis: How environ
mental mycobacteria may predetermine the protective efficacy of BCG. Tubercle
62, 55-62.
I
60
The species of mycobacteria
4
’ - nonphotochromogcnic
nov„ nom. rev. International
sp. i-......
Diagnostic
mycobacteriology
mycobacteria, including Mycobactenum
Journal of Systematic Bacteriology 31, 247-58.
nov., nom. rev.: a lung pathogen.
Tsukamura, M. 1982: Mycobacterium shimoidei sp.
which causes shin
infection. Microbiol. Immunol. 26, 951 5.
Numerical analysis of rapidly
I.,.•/»—* “'"1%Ji’™
Bacteriology 33,460-9.
Echemendia-Font, M. 1971:
i^"eu’va especie den.ro de ia micobacterias. Bote,in
de Higiene EPide^l0^ia^1y\ et al 1991; Clinical disease, drug suscepti-
within Group IU. Americnn Review
numerical analy-
and
marinum. Journal of General Microbiology 109. 3 9 2 ■ co.operalive numerical
^^oi ^h^Xw^g mycobacteria. Journal of General
Microbiology 66,255-71.
£ 1971. A new mycobacterium species:
r^'Mi^ogica AcLmicae Scieniiarun,
Diagnostic mycobacteriology is a complex and technically demanding branch
of medical microbiology. The organization of diagnostic services and the stan
dard technical procedures have been reviewed in detail elsewhere (Collins et
al., 1985). The development of DNA-based diagnostic techniques such as
DNApmhes and the pqlymerase_£haiiLreaction could revolutionize diagnosdT^obacteriology although at the time of writing there are probtans oi
sensitivity, specificity, cross-contamination andcftsl The aim ot this chapter
is toenable the clinician to make the most of the available services and to out
line the principles of the newer technologies.
The examination of clinical specimens suspected of containing mycobac
teria by standard methods consists of four steps:
1. microscopical examination;
2. isolation of mycobacteria in culture;
3. identification of the organism;
4. drug susceptibility testing (where relevant).
In addition, reference laboratories play an important role in the collection of
epidemiological information, staff training, quality control and research as
wdU? giving clinical andlechnical advice. In many countries, laboratories
are arranged in a hierarchical fashion. Microscopy and culture are performed
in peripheral laboratories, larger centres identify Mycobactenum tuberculosis
while the identification of other species as well as drug susceptibility testing
are usually performed in designated reference centres. In New York btate,
USA, an additional component to the laboratory service is the ‘fast-track pro
gramme’ which uses modem technology and focuses on potentially hig y
infectious patients (Salfinger and Pfyffer, 1994). In poorer countries, empha
sis is placed on microscopy services so that open or infectious cases ot
tuberculosis are detected and treated.
type of tubercle bacillus. Medical Research Council
Laboratory safety
SUE.'7=*.=“;x‘"7 £.=
Bacteriology 23, 182-3.
The handling of specimens and cultures exposes laboratory staff to a serious
risk of infection. Workers in autopsy rooms are at particular risk. Safety
measures are essential and, in some countries, mandatory. Tly656 mclu e^on
tainment laboratories, approved safety cabinets and centrifuges, protective
cfothing, hand basins and facilities for the safe disposal of contaminated waste
(World Health Organization, 1993). AlljtafLshould be vaccinated with the
Bacille Calmette-Guerin (BCG) vaccine, unless they are tuberculin positive
Specimens 63
62 Diagnostic mycobacteriology
or have a BCG scar, and must have medical supervisipn as determined by
10CHeal°thCand safety measures are no substitute for adequate training of staff
in the basic technical procedures, their constani^ar^si_pfpotentia]Lh|Z
ards and their familiarization with local safety rules. For a fuH account ol lab
oSo"ty see Collins (1993), Collins et al. (1995) and World Health
OrClmiciansn haveaduty to consider the welfare of the laboratory staff by sub
mitting specimens in a safe condition and by ensuring that sP^c"nen5^
ers are not contaminated on the outside. Laboratory stalf must be informed if
the specimen comes from a patient known to be infected. with a ^ngcrou
nathoeen such as hepatitis B or the human immunodeficiency virus (H V)
Plastic bags containing tracheal catheters, bronchoscope traps and other
obiects contaminated with sputum are particularly hazardous and should e
disposed of immediately without being opened. Request forms should neve
be p?aced in'the same bag as the specimen. Plastic bags wtth separate pockets
for the request forms are available.
The collection of specimens
The first and most important, steps in any microbiological investigation are
the
prompt delivery of suitable spectmens. Much valuabLe
nfnSr don is joshanTtoieh wasted, by the submtsston of inadequate or
mislatelled specimens or by delays in their delivery to the laboratory.
It is particularly important to collect all specimens into sterile containers,
is a common misassumption that, as specimens for mycobacterial mvesttgatinns are ‘decontaminated’ before culture, the cleanliness of the container is
VessS imporunt11^^ in the case of other bacteriological examinations.
Unfortunately, unsterilized_containers may_be contiumnated wdh environ
mental mycobac^There have been a number of
apparent^sease due to environmental mycobacteria as a result ot coHec. o
soutum or urine in clean but unsterile containers (Collins ef a I 1984)
Mycobacterium xenopi is a common culprit as, being thermophilic, it sui vi ves
in-hospital hot water systems and contaminates sputum pots that are washed
under the hot tap. Equipment used to collect specimens may likewise be contamtnated simdar pseGdoepidemics have been traced to the use of inade
quately sterilized endoscope^(Gubier et al., 1992).
Specimens
Soutum The most suitable containers for sputum are wide-mouthed, screw
capped disposable plastic pots, into which sputum may be directly expec
^Most sputum specimens submitted for examination are Lsp^samples
taken at clinics, but it is preferable to obtain at_leastjhree early-morpm^
specimens Sputum specimens are inevitably contaminated with non-aadfast organisms which multiply rapidly. Specimens should therefore
conveyed to the laboratory as quickly as possible. If delays in transport are
unavoidable, sputunrrnay.hectored for up to one week at 0°C. If specimens
are to be transported for long distances in hot climates and (as is often the
case) refrigerated transport is not available, sputum may be preserved by the
addition of an equal volume of 1 per cent cetyl pyridinium chloridejn 2 per
cent w/v saline.
Other respiratory tract specimens. Sputum is by far the best material for the
diagnosis of pulmonary mycobacterial disease and all efforts should be made
to obtain a specimen. Laryngeal_swabbing and gastric aspiration are relatively
ineffectual means of obtaining material suitable for culture, as well as being
unpleasant for the patient. A more satisfactory alternative, if the equipment is
available, is the use of the fibrepptic bronchoscope to take biopsies of radiologically visible lesions, to sample material on the bronchial walls by means
of a small extendable brush, or to rinse out sections of the bronchial tree with
saline (broncho-alveolar lavage). After use, the bronchoscope is sterilized
with a 2 per cent solution of glutaraldehyde.
Laryngeal swabs are specially prepared for the purpose: they consist of stiff
wires bent at one end at an angle of 35° and tipped with cotton wool. Some
laboratories issue swabs lipped with alginate wool which can subsequently be
dissolved so as to liberate the mycobacteria. Swabbing should only be done
by staff trained in the technique: blind swabbm^of the pharynx is useless. The
operator should wear a visor as the patients inevitably cough violently during
Gastric aspiration is performed in the early morning before food or drink is
taken. The patient is requested to cough and swallow several times before the
stomach contents are aspirated through a naso.gasiriclube. The aspirate should
be transported without delay to the laboratory or, if this is not possible, it
should be neutralized with sodium hydroxide.
Urine Three early-morning midstream specimens_should be obtained in 28
ml alass or plastic (‘Universal’) containers. Some laboratories isolate myco
bacteria by membrane fillration of larger quantities of urine, such as total
early-morning specimens.
Other fluids. Cerebrospinal fluid (CSF).and pus should be placed directly in
sterile containers and sent to the laboratory without delay Examination of
CSF is an emergency investigation and the laboratory staff should be con
tacted by the clinician. Specimensjif plcuraLpericardial and peritoneal fluids
may contain enough fibrinogen to cause them to coagulate. This is pre
ventable by adding sterile sodium citrate to the specimen. Alternatively, such
fluids may be added directly to an equal quantity of a double-strength liquid
medium such as Kirchner or Middlebrook broth.
Tissues. Tissue is much more suitable for culture than necrotic material or pus,
probably because the latter contain free fatty acids that are toxic to myco
bacteria. For this reason the Urgest of a group of excised lymph nodes may not
be the best for culture as it may be the most necrotic. Biopsies should be
examined histologically as well as bacteriologically. This requires two speci
mens - one preserved in formalin for histology and the other one freshJpr
bacteriology.
I
Culture media 65
64 Diagnostic mycobacteriology
Blood and bone marrow. In view of the increasing frequency of HIV-related
disseminated mycobacterial disease, it is often necessary to cxamine_blQ£)d
and bone marrow. Radiometri£lechniquesJ‘BactGck.see below) are ideal for
the examination of "these materials which are inoculated directly into the
medium according to the manufacturer’s instructions. Alternative culture
methods include inoculating blood or bone marrow directly into liquid media
or centrifuging anticoagulated blood lysed with sodium deoxycholate and
inoculating solid media with the centrifuged deposit.
Faeces. Another consequence of HIV-related mycobacterial disease is the
need to culture faeces for mycobacteria. Samples of 1-2 g should be submit
ted in sterile containers. Faeces are unsuitable for the diagnosis of tubercu
losis in HIV-seronegative patients.
Microscopy
As the isolatioikoLAL tuberculosis and most other pathogenic mycobacteria
by standard cultural methods takes several weeks, the use of microscopy to
reach a preliminary diagnosis is of great importance. In addition, microscopy
of sputum is of great value in the detection of open or infectious cases of
tuberculosis. It is well established that if no acid-fast bacilli are seen in spu
tum on a standard, yet competently performed, examination, the patient is
unlikely to be infectious. (This may not be the case with HIV-seropositive
patients.) Thus the establishmenLof-good sputum microscopy services is of
prime importance in developing countries where the first priority in tubercu
losis control is the detection and treatment of the open cases.
Sputum is examined either directly or alter liquefaction and centrifugation
(see page 66). Direct examination is performed by selecting a purulentlooking portion of sputum and spreading it thinly on a glass slide with a
bacteriological loop or a wooden throat swab stick. If centrifuged deposits are
examined, great care must be taken to make sure that the reagents and diluents
are not contaminated with environmental mycobacteria in order to avoid
false-positive smears. There is evidence that concentration of bacilli in spu
tum by use of the cytocentrifuge increases the sensitivity of microscopy to
that of culture on solid media (Saceanu et al., 1993; Fodor, 1995).
Smears are stained by the Ziehl-Neelsen (ZN) method, or by one of its
various modifications. In the traditional ZN method heat-fixed slides are
flooded with carbol fuchsin (a phenol/water solution of basic fuchsin) and
heated until steam rises. Boiling must be avoided as this dislodges the smear
from the slide. After five minutes the slide is washed under the tap and then
flooded with a dilute mineral acid such as 3 per cent sulphuric or hydrochloric
acid. Some workers use acidified alcohol rather than acid alone as this tends
to give a cleaner film. It does not, contrary to popular belief, distinguish tuber
cle bacilli from other mycobacteria.
.
After five minutes the slide is washed again and a counterstain is applied.
The counterstain may be green or blue, depending on the microscopist s pref
erence: red-blue colour-blind workers may use a yellow stain such as picric
acid. Various ‘cold’ ZN staining techniques have been introduced but are not
An alternative to the ZN method is staining with auramine or rhodamine (or
both) for fluorescence microscopy (FM). Although the equipment is more
expensive, FM is much less tiring to the technical stall. Smears arc examined
at a lower magnification, thereby increasing the chance of detecting low num
bers of acid-fast bacilli. Some authorities consider that fluorescent staining is
less specific than ZN staining and advise that all smears found to be positive
by FM should be confirmed by the ZN method.
u u
Biopsies may be homogenized by grinding with sand in a Griffith tube or
examined histologically by use of a modified ZN staining procedure such as
that used by the US Armed Forces Institute of Pathology (see Ridley, 1988).
Urines, cerebrospinal fluids and other fluids are centrifuged and the deposits
are stained.
Culture media
The tubercle bacillus was originally isolated on heat-coagulated serum.
Nowadays the most popular media contain egg and are solidified by heating at
80°C for one hour, a process termed inspissation. The widely used
Lowenstein-Jensen medium contains eggs, asparagine, glycerol and some
mineral salts. Stonebrink’s medium is very similar to the former but contains
sodium pyruvate instead of glycerol and is therefore suitable for the isolation
of bovine tubercle bacilli. Egg-based media usually contain a dye such as
malachite green which, in addition to being inhibitory to certain contaminat
ing bacteria, gives a better background colour against which colonies of
mycobacteria are more clearly seen. The nutritional role of egg is uncertain.
The yolk may provide lipid precursors but it is doubtful if the albumin protein
is utilized. It is more likely that the role of the protein is to absorb tree fatty
acids which are toxic for mycobacteria, although they stimulate growth if pre
sent in small quantities.
There are a number of clear broths or agar-based solid media but these are
used more for drug susceptibility testing and research work than for primary
isolation of strains from clinical specimens. These include MiddlebrookDubos 7H9 broth and 7H10 agar, Sauton’s medium and Kirchner’s broth.
For details see Collins et al. (1985). Liquid media are used in the radiometric
culture technique (see below). A commercially available system (Septichek
AFB system, Becton Dickinson) combining a liquid medium and a solid
medium on a slide, permits detection of mycobacterial growth more rapidly
than by standard culture but not as rapidly as by radiometry (Salfinger et al..
The above media support the growth of almost all mycobacteria (except M.
leprae and a few other non-cultivable species) but a few species require
special media. The use of pyruvate instead of glycerol for the isolation of
bovine tubercle bacilli has already been mentioned. Mycobacterium para
tuberculosis and some strains of M. avium require media supplemented with
mycobactin (see page 26) for growth. Such media are also preferable to stan
dard media for the isolation of M. avium from specimens likely to contain
only very few bacilli. Media containing iron supplements (haem or ferric
ammonium citrate) are required for the isolation of M. haemophilum.
Radiometric methods 67
66 Diagnostic mycobacteriology
on blood agar overnight to check for the need for a decontamination procedure.
Decontamination of specimens
hhi
nated - a fact that must be accepted by clinicians.
•
.Soff
The choice of decontaminating agent depend on the spec,
Incubation and reading the cultures
There is a considerable variation in the temperature range of growth of the
mycobacterial species - a feature utilized for identification purposes. Most
mycobacteria grow at 35-37qC, but three species associated with skin disease,
namely M. marinum, M. haemophilum and M.
grow at^a low"
perature. Thus all material from skin lesions should be ,n“bat^ a‘33 ^aS
well as at a higher temperature. Mycobacterium chelonae may fail to grow. a
37‘c on primty isolation but grows at 35“C. Mycobactena are aerobic but
he growth is enhanced by an atmosphere of 5-10 per cent CO m. air. A CO
incubators are cosily to obtain and maintain, they are rarely used in chmcal
Pr Tte slopes should be examined weekly for at least 8 weeks and preferably
for up to 12 weeks as some species, such as M. xenopi and M. malmoens ,
may Fake this time to appear on primary isolation An even longer11}“ba‘10 ■
up to 14 weeks, is required in veterinary practice for the isolation ot M.
sputum a liquefying agent sue
decontaminating
reagent tolid concentration of acid-fast b.aci“*Sputum is shaken
using neutral red as an mdicator.
techniques in which
^“co'ntminafion is usually indicated by a softening or discolouration of the
media Mycobacterial colonies can usually, with expenence, be distinguished
from those of contaminating bacteria but confirmation should always be made
by Ziehl-Neelsen staining.
Animal inoculation for the isolation of
mycobacteria
Guinea-pig inoculation was once a popular way of diagnosing tuberculosis
but it is now obsolete. It has been clearly demonstrated that the use of this ani
mal offers no advantages over in vitro culture (Fallen, 198/).
Radiometric methods
In view of the slow rate of growth of tubercle bacilli and most other patho-
based on the release of radioactive CO, from a labelled Pr<jcursor
metabolism. The released gas is detected by periodic sampling of he a'r Pa“
over the medium: this is done automatically in a commercially available
instrument (Bactec 460/TB, Becton Dickinson). The specimen is added to a
bottle of broth containing [,4C]-palmitic acid and a cocktail of antimicrobial
iPents such as PANTA (see above), to inhibit growth of any organism other
EX'; Growth of mycobacteria may be detectable within two or
Drug susceptibility testing
68
69
Diagnostic mycobacteriology
three days, although most studies show that the isolation rate is no higher than
by standard culture. A few strains of M. tuberculosis grow on conventional
solid media but not in the radiometric system so both culture methods shoulc
^Radiometry is also used for rapid drug susceptibility testing by incorporat
ing antituberculosis drugs in the media. Although it is undoubtedly more rapid
than conventional techniques, it is more cosily and this restricts its use in rou
tine diagnostic services in many countries. For further details ot radiomen
methods, see Heifets (1986).
Identification of mycobacteria
By far the most frequently isolated mycobacterium in clinical practice is M.
tuberculosis. The first step in identification is therefore to determine whethei
or not an isolate is of this species or the closely related M. bovis and
africanum. This can be done in several ways; tor examp e, the lollowing
properties together clearly distinguish these strains from all other mycobac
teria: slow growth rate, no pigment produced in the light or dark, no growth at
25°C and no growth on egg media containing 500 mg/1 of p-mtrobenzoic acid
or p-nitro-a-acetylamino-p-propriophenone (NAP).
Some workers identify the human tubercle bacillus by Us producuon o
large quantities of niacin, detectable by a simple chemical test (Konno, 1956
or by the use of commercially available lest strips (Difco). Reliance shoulc
not be placed on this test alone as, in addition to bovine strains, a lew human
strains are negative. Conversely, positive reactions are given by M. simiae and
by a few strains of M. avium and the rapidly growing pathogen M. chelonae.
Nucleic acid probes for the rapid identification of the M. tuberculosis com
plex and, specifically, of M. tuberculosis are commercially available (see page
74).
Typing tubercle bacilli
The tubercle bacilli that cause human disease are M. tuberculosis, M. bovis
and M. africanum. In addition BCG occasionally causes localized or even
widespread disease and may need to be distinguished from the virulent types.
The subdivision of the virulent tubercle bacilli is for epidemiological pur
poses only as the clinical management of all these types is identical, except
that bovine strains are naturally resistant to the antituberculosis drug pyrazinamide. A widely used typing scheme for these organisms is described on
page 43 and the technical methods have been published by the World Healt i
Organization (Grange and Yates, 1995).
Identification of other mycobacterial species
There is no universally accepted protocol for the identification of mycobac
teria other than the tubercle bacilli. Ideally each strain thought to be the cause
of disease should be identified al the species level and, in some cases, at the
subspecies level so as to increase our knowledge of the types of mycobacteria
that cause disease, and to determine the most appropriate therapy. Often,
though, facilities and finance prevent such a thorough investigation.
Some laboratories are in the fortunate position of being able to perform a
wide range of tests on each isolate and possibly to use highly discriminatory
test systems, including nucleic acid probes and ribotyping. Others must be
content with using a few simple tests which will identify most species that are
encountered in clinical practice. Marks (1976) divided clinical isolates into 15
groups according to growth at 25, 37, 42 and 45°C, pigment production oxy
gen preference and hydrolysis of Tween 80. Additional simple and useful less
include detection of nitratase and arylsulphatase activity and reduction of tel
lurite. Glycosidase activity is also of value, especially a-L-fucosidase activity
which is particularly strong in M. marinum and also distinguishes M.szulgai
from other slowly growing scotochromogens. The use of these and other cul
tural and biochemical tests for the identification of most of the slowly grow
ing mycobacteria is shown in Table 4.1.
Rapidly growing species - those that give a good growth on subculture
from a small inoculum on Lowenstein-Jensen medium within seven days
may be identified by a range of enzymic activities and substrate utilizations.
In this respect, utilization of citrate, ammonia production from allantoin and
acid production from mannitol, inositol and xylose and are of particular value,
as shown in Table 4.2. From the clinical point of view the important rapid
growers are M. chelonae and M.fortuitum as, with very rare exceptions these
are the only human pathogens in this group. These are distinguished from
other rapid growers by their lack of pigment, strong arylsulphatase activity
and limited saccharolytic activity, and from each other by the nitratase test and
other properties shown in Table 3.5.
.
More sophisticated identification tests include characterization of mycosides (sec page 21) by lipid chromatography, immunodiffusion analysis of
cytoplasmic antigens, protein electrophoresis, gas-liquid chromatography,
pyrolysis mass spectroscopy and ribotyping (see page 29). Nucleic acid
probes for rapid identification of some of the commoner mycobacteria iso
lated in clinical laboratories are commercially available (see page 74).
Drug susceptibility testing
As outlined later (see Chapter 10), drug-resistant mutants continuously arise
at a low rate in any mycobacterial population. Any culture will therefore
inevitably contain a few such mutants. The purpose of susceptibility testing is
to determine whether the great majority of bacilli in the culture are sensitive
to the antituberculosis drugs currently in use. In other words, susceptibility
tests are designed to inform the clinician whether or not an isolate is as sus
ceptible to a given drug as other known sensitive strains.
There are four major techniques for susceptibility testing: the resistance
ratio method, the absolute concentration method, the proportion method and
radiometry. Testing may be direct or indirect, i.e. performed on the original
specimen or on a subculture respectively.
Table 4.1 Properties of the slowly growing mycobacteria
Arylsulphatase Tween 80 Tellurite a-L-fucosidase Urease Acid
J.
Nitrate
Pigmentation Growth
at (T")
0C)^
phosphatase
hydrolysis reduction
20 25 33 42 44 reductase
M. tuberculosis
M. bovis
M. kansasii
M. marinum
M. asiaticum
M. simiae
M. scrofulaceum
M. szulgai
M. gordonae
M. avium-intracellulare
M. malmoense
M. ulcerans
M. xenopi
M. haemophilum
M. terrae
M. triviale
M. nonchromogenicum
N
N
P
P
P
P/N
S
+
V
+
v
+
s
s
N
N
Ny
Ny
N
N
N
N
v
+
v
+
(+)
(+)
V
+
(+)
+
(+)
0
+
0
0
0
0
0
0
+
V
V
V
- - scotochromogen; Ny = light lemon-yellow colour; + = positive reaction in >85 per cent of strains;
scotochromogen; Ny = light lemon-yellow co,°ur.
N = nonchromogen; Pin= As
photochromogen;
S - (+) = weak or late reaction; v = variable growth; 0 = Itrnted or no data
per cent o’t strains;
Table 4.2 Properties of some rapidly growing mycobacteria
Pigment
M. fortuitum type A
M. fortuitum type B
M. fortuitum type C
M. chelonae abscessus
M. chelonae chelonae
M. smegmatis
M. phlei
M. diernhoferi
M. gilvum
M. duvalii
M. flavescens
M. vaccae
Growth
at 45°C
Growth after Arylsulphatase Nitrate
4 hours at
(3 days)
reductase
60°C
Citrate
utilization
Allantoinase Acid from Acid from Acid from
mannitol inositol
xylose
+
+
+
v
+
+
positive; - = negative; v = variable result
+
V
V
V
+
Laboratory methods in leprosy
73
Diagnostic mycobacteriology
72
The resistance ratio is determined by inoculating standardized suspensions
of the test strain and a number of known sensitive strains on to media con
taining doubling dilutions of the drug. After incubation, the endpoint for each
strain (i.e. the slope with 20 or fewer colonies) is determined. Test strains are
then compared with the average or ‘modal’ resistance of the set of known sen
sitive strains. If the endpoint of test and controls is equal, the strain has a
resistance ratio of 1. As doubling dilutions of drugs are used, 1,2 or 3 tube dif
ferences in the endpoints of test and control strains give resistance ratios of 2,
4 and 8, respectively. Strains with resistance ratios of 4 or more are reported
as resistant. Examples are shown in Table 4.3.
The absolute concentration method is very similar to the resistance ratio
method, being based on a titration of the test strain, along with adequate con
trol strains, on slopes of media containing known quantities of the drug in
doubling dilutions. The difference is that the results are expressed in the actual
endpoint concentration of drug. In practice, the activity of the drug may be
less than its concentration in the medium owing to its denaturation during
medium preparation. Accordingly, this method has no real advantage over t e
resistance ratio method.
.
__ . ,
In the proportion method the number of colonies growing from a standard
inoculum on a drug-containing medium is compared with the colony count
from the same sized inoculum on a drug-free medium. A strain is considered
resistant to a given concentration of drug if the number of colonies growing
on the drug-containing medium is 1 per cent or more of the number growing
Collins et al. (1984, 1995) and Vareldzis et al. (1993) and for radiometric
methods see Heifets (1991) and Heifets et al. (1993).
Pyrazinamide susceptibility tests pose a particular problem as this drug is
only active in acid media and there is a narrow pH range, around 5.2, at which
the drug is active and the bacilli are able to grow. A reliable technique was
described by Marks (1964) and modified by Yates (see Collins et al., 1985;
Grange and Yates, 1995). Pyrazinamide-resistant strains lack the enzyme,
pvrazinamidase, required for conversion of this drug to its active metabolite.
Thus, detection of this enzyme activity may prove to be a simple alternative to
formal susceptibility testing, but this has not been adequately investigated.
Laboratory reports
A report of the microscopic examination may be issued soon after receipt of
the specimen. As tubercle bacilli cannot be distinguished from other myco
bacteria on the basis of microscopy the initial report should merely state that
acid-fast bacilli are present. Some laboratories also report the number ot
bacilli seen. A scale widely used in Europe for reporting smear microscopy is
shown in Table 4.4.
Table 4.4 Sputum smear microscopy: a scheme for reporting the number of acid-fast
bacilli seen in high power fields (Collins et al.. 1995)
°nThelhUreefrnonmradiometric methods give essentially similar results. The
Number of bacilli seen
Report
resistance ratio method is used in Great Britain while the proportion method
is widely used in the USA. More recently, radiometry (see page 67) has
proved to be a rapid and reliable method for determining susceptibility io the
first-line drugs and its rapidity justifies its higher cost.
For details of the standard techniques for drug susceptibility testing sec
None in 300 fields
1-2 per 300 fields
1-10 per 100 fields
1-10 per 10 fields
1-10 per field
>10 per field
Negative
( ) Repeat test
Table 4.3 Examples of the resistance ratio method
Modal resistance
Test strain no.
1
2
3
4
5
6
7
8
Resistance
ratio
Increasing drug concentration
(tube number)
12
3
4
5
6
C
D
O
O
O
O
C
C
C
C
C
C
C
C
C
D
C
C
C
o
o
o
o
o
O
O
O
O
o
o
o
o
o
o
o
c
c
c
D
C
D
C
C
C
D
+
c
o
c
D
C
D
c = confluent growth; D = numerous discrete colonies;
colonies
1
1
2
4
4
8
8
8+
20-100 colonies; O = <20
+++
++++
Isolation of a mycobacterium in culture usually takes from two to five
weeks and at that stage it should be possible to say with a high degree ot cer
tainty whether or not the isolate is a member of the M. tuberculosis^complex.
A firm identification of the strain, together with results of susceptibility test
ing, is available after a further three or four weeks. Thus susceptibility results
are’usually available about two months after submission of the specimen,
unless direct testing or radiometric methods are used.
Laboratory methods in leprosy
As M. leprae cannot be cultured in vitro, routine investigations are limited to
the microscopic demonstration of the bacilli in clinical specimens although
the polymerase chain reaction (PCR) will become increasingly available tor
the rapid detection of this bacillus. The usual specimens for the microscopic
diagnosis of leprosy are slit-skin smears and nasal scrapings which are
examined for the presence of acid-fast bacilli. These diagnostic procedures
References
74
75
Diagnostic mycobacteriology
m. wt
kb
are described in Chapter 7. Leprosy bacilli may be propagated in the footpads
of mice and, by incorporating drugs in the animals’ food, susceptibility testing
may be performed. This is a laborious procedure that is only carried out in a
limited number of reference and research centres. Alternative techniques for
determining drug susceptibility of M. leprae have been developed and are
mostly based on the uptake of radio-labelled substrates (reviewed by Grange,
1991) Radiometric techniques (Bactec system) may also be used to determine
in vitro drug susceptibility of M. leprae (Tomioka et al., 1992) although large
numbers of bacilli, 5x lOGo 1 x 10s. are required. The PCR has also been used
to detect M. leprae in tissues and preliminary observations indicate that it is
sensitive and specific (Jamal et al., 1994). In addition, rifampicin resistance
may be rapidly detected by single-strand conformational polymorphism (see
page 30) of the PCR-amplified rpoB gene of M. leprae.
10
5
2
0.5
Laboratory methods based on DNA technology
The three techniques that are increasingly being used in diagnostic labora
tories are PCR to detect mycobacterial DNA in clinical specimens, nucleic
acid probes to identify cultures, and restriction-fragmenU^nglh polymor
phism (RFLP) analysis (DNA fingerprinting) to compare strains lor epidemi
ological purposes. The principles of these methods are described in Chapter 2
The diagnosis of tuberculosis using PCRh,as been extensively investigate
but it has notproved as sensitive_oras. specific as originally hoped. A blind
comparison study by seven laboratories all experienced in molecular tech
nology revealed a wide diversity in sensitivity and specificity (Noordhoek et
al. 1994) and serious doubts.have been raised as to the usefulness ol PCR, m
the form available in 1995, as a diagnostic tool (Grosset and Mouton, 1995).
Nevertheless, much research activity is being undertaken and innovative
modifications of the PCR in commercially available kit form are becoming
available. One useful innovation is the amplification of specific rRNA, o
which there are about 2000 copies in each cell rather than just one or up to 20
in the case of insertion sequences) of a specific DNA sequence in the genome.
Such amplification may be achieved without the need lor thermal cycling, and
test kits which are commercially available, detect mycobacteria in clinical
specimens with a high degree of sensitivity and specificity (Bodmer et al.,
Fig. 4.1 /S6110 RFLP analysis of a collection of M. tuberculosis isolates (courtesy of M.
Yates,’Dulwich PHLS) including both classical European isolates and Asian strains. DNA
was digested with Pvull. electrophoresed on 1% agarose gels. Southern blotted and
hybridized with a MP labelled probe for IS6110 and then autoradiographed. Some of the
asian strains were not polymorphic, with only a few bands, resulting from low copy
numbers of IS6110. Courtesy of Dr P. Butcher. St. George’s Hospital. London
The technique of DNA^fingsrprinting has found several applications in
epidemiological research (Stoker, 1994). Il assistsjn detecting the source of
infection, especially in ‘explosive’ HIV-related mini-epidemics and in distinguishing between exogenousjginfection and endogenous reactivation. It is
also”useful for detecting instances of laboratory cross-contamination of
cultures. An example of the results of DNA fingerprinting is shown in
Fig-4-1Although such molecular techniques are not widely available, some coun
tries have centres that perform PCR and DNA fingerprinting for specific pur
poses such as rapid diagnosis of tuberculous meningitis and investigations of
outbreaks of disease. For further details on the clinical application of these
newer techniques see Salfinger and Morris (1994) and Salfinger and Pfyffer
(1994).
'^Nticleic acid-probes with non-isotopic detection systems are available tor
the M. tuberculosis complex, M. tuberculosis, M. avium, M. intrace11 ula re, the
M avium complex, M. kansasii and M. gordonae and doubtless others wil
become available. These can be used to identify cultures from conventional
media and from radiometric vials (Salfinger and Pfyffer, 1994). The accuracy
of these is very high but not absolute. Cultures of M. celatum and M.terrae
have, for example, been misidentified as M. tuberculosis. Thus identification
by use of probes must be followed up by conventional confirmatory tests.
Molecular techniques may eventually replace the time-consuming rug
susceptibility tests.. Techniques are available for detecting mutations
responsible for rifampicin resistance in the PCR-amplified rpoB gene (sue
References
Bodmer, T., Gurtner, A., Schopfer, K. and Matter, L. 1994: Screening of respiratory
tract specimens for the presence of Mycobacterium tuberulosis by use of the GenProbe Amplified Mycobacterium Direct Test. Journal of Clinical Microbiology 32,
!483-7.
J
n
Collins C H. 1993: Laboratory acquired infections. 3rd ed. London: Butterworths.
Collins’, C.H., Grange, J.M. and Yates, M.D. 1984: Mycobacteria in water. Journal oj
Applied Bacteriology 51,
Collins, C.H., Grange, J.M. and Yates, M.D. 1985: Organization and practice m
.
l .......inric
T onrion* Riittprwnrf
References
76
77
Diagnostic mycobacteriology
Collins C.H., Lyne, P. and Grange, J.M. 1995: Microbiological methods. 7th ed.
FotaTOni995‘ DZ^fZobaeteria in spututn spears prepared by cytocentrifuG±enj "-"on ofXg^ce in Mycobacterium lepraethe
Sgn treatment regimens for leprosy. In Heifets L. (ed.) Drug ^scep tb Itty .tn
the chemotherapy of mycobacterial infections. Boca Raton. CRC Press, Inc.
rrnna.’61?^’ and Yates MD. 1995: Guidelines for speciation within the
GMy^bacmnum^utoc^
Geneva: Wodd Health Organ,zalton.
Gro™?7:md Moutom Y. 1995: Is PCR a useftil tool for the diagnosis of tuberculosis
G'Cler’jGH^Samn^r
1992: Pseudoepidemic of nonG“Xu?ous mycobfcteria due to a contaminated bronchoscope cleamng machine — — ^"^mycobacteri-
HeSy'"99f:X^^^^^
etal. 1993: Radiometric broth macrodilu-
Ja™1I"nysanM.ReHSMM^M994: A colorimetric PCR method for the
detection of M. leprae in biopsies from leprosy patients. International Journal of
Ko^nno^K ll’sb: New chemical method to differentiate human type tubercle bacilli
^«efcr the estimation of pyrazmamide sensithe examination of tubercle bacilli and other mycobacMitchiso^D ^’llem B.W. and Manickavasagar, D. 1983: Selective Kirschner
medium in the culture of specimens other than sputum for mycobacteria. Journal of
G. et al. 1994: Sensitivity and specificity of
N°PC^ t&tectiorfof Mycobacterium tuberculosis' a blind comparison study among
n nrHhnpk^T^olk6’/
cases of mycobacterial infection - does it aid diagnosis and treatment? Tubercle 68,
Ridley8 D.S. 1988: Pathogenesis of leprosy and related diseases. London: Wright.
961 79.
inon- Cnmnnrison between the MB
Check system, radiometric and conventional methods for recovery of mycobac
teria. Journal of Microbiological Methods 12, 97-100.
Stoker, N. 1994: Tuberculosis in a changing world. British Medical Journal
Tomioka^H., Saito, H. and Sato, K. 1992: Evaluation of BACTEC 460 TB system for
measurement of in vitro ^-Mycobacterium leprae activity of various antimicro^93:
resistant tuberculosis. Geneva: World Health Organization. ^0^3.171.
World Health Organization. 1993; Laboratory biosafety manual, 2nd. edn. Geneva.
World Health Organization.
Tuberculosis 205
10
Therapy of mycobacterial
disease
The antimycobacterial agents
These are divisible into those substances that are only used lor treating
mycobacterial disease and those that have a much wider use. As a general
rule, the modes of action of the drugs used specifically for treating mycobac
terial disease are much less well understood than those of the agents also used
for other purposes. For details of the pharmacokinetics and mode of action of
the conventional antimycobacterial drugs see Winder (1982) and Bartmann
(1988).
Tuberculosis
From time immemorial, the search for a cure for tuberculosis has been a major
preoccupation of the medical profession. Indeed, it is salutary to recall that
effective drugs have been available for only a few decades. Before the
causative agent of tuberculosis was discovered in 1882, huge numbers of
remedies were advocated: many were bizarre, most were unpleasant and vir
tually all were useless. Although Sir William Buchan’s advice that the patient
should suck woman’s milk directly from the breast contrasts sharply m
acceptability with John of Gaddeston’s prescription of a mixture of pigeon’s
dung and weasel’s blood, it was probably equally ineffective. Perhaps the only
remedy of some value was_cod Ji ver oil> first used by Percival in 1770. This,
owing to its high vitamin D content, may have exerted a beneficial effect in
some cases of lupus vulgaris and other superficial tuberculous although it may
have Ted to progression of pulmonary disease.
For about four decades after Koch’s discovery of the tubercle bacillus,
many workers, including Koch himself, attempted to develop immunothera
peutic regimens. Sphalinger (1922), for example, raised antisera in horses
(Irish hunters) with which he treated a few patients, including President Nehru
of India, with apparent success. The era of immunotherapy ended with the dis
covery of effective antituberculosis drugs, commencing with streptomycin by
Waksman in 1948, but immunotherapy is once again set to play a major role
in the control of tuberculosis (see page 220).
The importance of adequate chemotherapy in the control of tuberculosis
has been well-established. Modem chemotherapeutic regimens are curative in
the majority of cases provided that the tubercle bacilli are susceptible to the
drugs and that a full course of therapy is taken. Ideally, treatment should be
themselves. For this reason, much research has gone into establishing the
shortest duration of therapy and smallest number of doses that are compatible
with a high cure rate.
Three other factors are of crucial importance to the success of chemother
apy as a component of tuberculosis control. First, effective drug regimens
must be prescribed. A study in India showed that many private general practi
tioners prescribed inadequate and expensive courses of therapy (Uplekar and
Shepard, 1991). Secondly, there must be a ready supply of drugs that are
known to be effective. The bioavailability of component drugs in some com
mercially available combination tablets is suboptimal (Fox, 1990); some
drugs are improperly stored or are time-expired and counterfeit drugs are mar
keted in some regions. Thirdly, patients should not have to pay for their drugs
or supervision of therapy.
Susceptibility and resistance to antituberculosis agents
In the absence of mutational resistance, tubercle bacilli are remarkably uni
form in their susceptibility to the antituberculosis drugs, although bovine
strains are naturally resistant to pyrazinamide. Mutational change to resis
tance occurs at a low and constant rate (Table 10.1), which varies from drug
to drug (David, 1970). Consequently, large populations of mycobacteria will
inevitably contain a few bacilli that are resistant to a given drug. The purpose
of drug susceptibility tests is not to detect these few mutants but to determine
whether the great majority of the bacilli are susceptible. In the absence of the
drug, the number of resistant mutants in a bacillary population remains low
but in the presence of the drug the susceptible bacilli are killed and the resis
tant ones become dominant.
Principles of chemotherapy
The aim of therapy is to destroy all viable bacilli, rather than merely to reduce
them to a very low level. This is not as straightforward as it might appear:
drugs that are fully bactericidal in vitro may not achieve this effect in vivo.
Mitchison (1985) has stressed the important difference between drugs that are
bacteriostatic, those that are bactericidal under permissive conditions, and
those that are capable of sterilizing lesions. Two further important properties
Table 10.1 Spontaneous mutation rate (per cell division) of Mycobacterium tuberculosis
to rifampicin, isoniazid, ethambutol and streptomycin. Data from David (1970)
Drug
Mutation rate
Rifampicin
Isoniazid
Ethambutol
2.3 x 10'°
2.6 x 10’
1.0 x 10'
3 Ox 10’
206
Tuberculosis
Therapy of mycobacterial disease
Table 10.2 The ability of antituberculosis drugs to sterilize lesions, reduce viable bacterial
population rapidly and prevent the emergence of drug resistance. Data from Mitchison
(1985)
Drug
Sterilizing
Early
bactericidal
Prevention of
drug resistance
Rifampicin
Pyrazinamide
Isoniazid
Ethambutol
Streptomycin
Thiacetazone
Good
Good
Fair
Poor
Poor
Poor
Fair
Poor
Good
Fair
Poor
Poor
Good
Poor
Good
Fair
Fair
Poor
207
in some short-course regimens, notably those used when drug resistance to one
or more of the other drugs is suspected or known or for retreatment of relaps
ing disease. Streptomycin has the disadvantage that it must be given by injec
tion, which may lead to transmission of the hepatitis B virus and the human
immunodeficiency virus (HIV) if needles and syringes are not sterilized.
Other antituberculosis drugs - ethionamide, prothionamide, thiacetazone,
viomycin, cycloserine and p-amino salicylic acid (PAS) - are of much lowei
efficacy although some, notably thiacetazone, are still used as first-line drugs
in certain countries on account of their low cost. These, and other, drugs are
also used to treat multidrug-resistant tuberculosis (see below).
Design of chemotherapeutic regimens for pulmonary
tuberculosis
of drugs are their ability to prevent the emergence of resistance to a second
drug, and their efficiency in destroying a large part of the bacterial population
rapidly, thereby greatly reducing the infectivity of the patient (Table 10.2).
The mycobacterial population in a patient is functionally divisible into
three groups:
1. freely dividing extracellular bacilli, mainly in the cavity walls;
2. slowly dividing bacilli within macrophages and in acidic, inflammatory
lesions, and
3. dormant and near-dormant bacilli, within cells and in firm caseous
material.
The most powerful sterilizing drug is rifampicin, being active against all three
groups. Isoniazid is the most effective agent for destroying the freely multi
plying extracellular bacilli, particularly those in the walls of cavities, but it is
not a good sterilizing drug as it has limited efficacy against the metabolically
less active intracellular bacilli which become relatively more frequent as
treatment progresses.
Pyrazinamide is effective at a low pH, such as is found within macrophages
and acidic, anoxic areas of inflammatory lesions, but it is ineffective against
organisms at a neutral or high pH. Thus it is not good at preventing emergence
of drug resistance as it only destroys part of the bacterial population.
Chemotherapy is divisible into three main phases. First, most of the freely
multiplying extracellular organisms (group 1) are destroyed. It is in this phase,
which lasts a week or two, that isoniazid has a major role. In the second phase,
lasting perhaps a month or two, the remaining bacilli, principally within
macrophages and inflammatory lesions, are killed by rifampicin and pyrazi
namide. Finally, only dormant or near-dormant bacilli remain and are eventu
ally killed by rifampicin. Despite the fact that isoniazid is usually only
bactericidal during the early phase of therapy, it is used throughout the period
of treatment as, by its bactericidal effect on replicating bacilli, it is very good
at preventing the emergence of drug resistance.
Ethambutol has some bactericidal action in the early stage of therapy but it
is not a sterilizing drug. Streptomycin is likewise not a sterilizing drug as it is
bactericidal in alkaline areas of the cavitv walls hut is ineffective at the low pH
Chemotherapeutic regimens are designed to cure the patient and prevent the
emergence of drug resistance. In view of the problem of mutation to resis
tance, it is essential to give at least two drugs to which the strain is susceptible
as the chance of two mutations occurring simultaneously in a single cell is
negligible. In practice, particularly in regions with a high prevalence ol drug
resistance, this involves giving three or more drugs.
In the early days of antituberculosis therapy, it was considered necessary to
givelETdrugs four times daily in order that inhibitory concentrations were
constantly maintained. As therapy then lasted for up to two years, as many as
3000 doses were required. Subsequently it was shown that it was more effec
tive to give the drugs daily and in some modern regimens they are given only
thrice or twice weekly. Indeed, tuberculosis could be cured in the majority of
cases with a mere 64 supervised doses.
For the reasons stated above, modern drug regimens are based on rifampicin
and isoniazid throughout and one or two additional drugs during an initial inten
sive phase. The regimen currently recommended by the World Health
Organization (1995) is known as Directly Observed Therapy, Short Course
(DOTS) and consists of rifampicin, isoniazid, pyrazinamide and ethambutol
daily for 2 months, and the first two drugs thrice weekly for a further 4 months.
Each dose is taken under direct supervision by a competent supervisor.
A number of alternative 6-month regimens are in use; all are based on
rifampicin, isoniazid and pyrazinamide with or without additional drugs and
all consist of an early intensive phase and a continuation phase (Table 10.3;
WHO, 1991; Centers for Disease Control, 1993). Some are daily and some are
intermittent, although the latter should be replaced by the DOTS regimen out
lined above whenever possible. Finally, although not recommended, 9- and
12-month regimens that do not contain rifampicin are still used in some
regions.
.
.
.
The advantage of using an early intensive phase of treatment is that it prob
ably cures a high proportion of patients so that there is less chance of the dis
ease relapsing if the patient absconds before completion of the less-intensive
continuation phase. The first- and second-line antituberculosis drugs and their
recommended doses are listed in Table 10.4. For further details see Davies et
Tuberculosis
208
Therapy of mycobacterial disease
209
well and the intrathecal administration of drugs in tuberculous meningms is
Table 10.3 Daily doses of antituberculosis drugs
Drug
Dose
Adults
Children
Rifampicin
450 mg if body wl <50 kg
600 mg if body wt >50 kg
Up to 20 mg/kg to
maximum of 600 mg
Isoniazid
200-300 mg
10 mg/kg
Pyrazinamide
1.5 g if body wt <50 kg
2.0 g if body wt >50 kg
35 mg/kg
Ethambutol"
15-25 mg/kg
As for adult
Streptomycin
750 mg if body wt <50 kg
750 mg if age >40 years
1 g if body wt >50 kg
20-30 mg/kg to
maximum of 1 g
Thiacetazone
150 mg
8-12 g (divided doses)
200 mg/kg
PAS
Ethionamide
750 mg (divided doses)
15 mg/kg
Cycloserine
500 mg (divided doses)
a Avsdrd
I:
;l.: ;.;l
4 mg/kg
required.
Not recommended
Therapy of multidrug-resistant tuberculosis
tested for visual acuity (see under Drug toxicity)
Table 10.4 Some addtonal drug regimens lor tuberculosis recommended by the World
Health Organization (1991)
Intensive phase: 2 months
Multidrug resistance is an increasing P.™b^
countries alike. The occuirence and ep
resistant (MDR) tubercle
renders standard
usTthe Advisory Council for the Elimination
Continuation phase: 4 months
Standard regimen: daily throughout
HRZ
For use when resistance to one drug is suspected: daily throughout
SHRZ
RR
EHRZ
Intermittent regimens: thrice weekly throughout
SHRZ
RR^
EHRZ
HR/
H = isoniazid; R = rifampicin; Z = pyrazinamide; S = streptomycin; E = ethambutol
Therapy of extrapulmonary tuberculosis
There is general agreement that the modern chemotherapeutic regimens dis
cussed above are suitable for the treatment of all types of non-pulmonary
tuberculosis, even life-threatening forms such as tuberculous meningitis.
There is less agreement over the duration of therapy: many physicians
dnue therapy for up to 12 or even 18 months. There is no real evidence that
this is necessary indeed, limited trials suggest that treatment for the standard
this is necessary.
mndprn rPO;mPn< penetrate tissues
Should receive four drugs r a’npiun son1
’ ^su
ibiHty tests are
given unde^direct observation by a
fess
210
Tuberculosis 211
Therapy of mycobacterial disease
antileprosy drug clofazimine is one of a group of related compounds origi
nally developed for treatment of tuberculosis and may again find application
for treating this disease. The fluroquinolones, such as ciprofloxacin, ofloxacin
and sparfloxacin and the new macrolides, clarithromycin and azithromycin,
have’been used with apparent success in a limited number of cases but con
trolled trials are required to establish their optimum use. As therapeutic regi
mens have to be ‘tailored’ for each patient on the basis of in vitro
susceptibility, comparative studies on various regimens are not easy.
Most of the recent experience in the management of MDR has been
obtained in the outbreaks in New York and other cities in the USA and is sum
marized by Dooley and Simone (1994), Morse (1994) and O’Brien (1994).
In view of the poor results of therapy of MDR tuberculosis, alternative
forms of treatment such as immunotherapy (see below) are urgently required.
disability and it has even been stated that steroids merely delay death and pro
Ions suffering (Escobar et al., 1975). A small study suggested that dexametha
sone reduced residual disability in both severe and mild/moderate cases but the
authors stated that no statistically meaningful conclusions could be drawn and
stressed the need for larger controlled studies (Kumarvelu et al., 1994).
Prednisolone, with an initial dose of 60 mg and a gradual reduction over an
11-week period, reduces mortality, effusions and subsequent constrictive scar
ring in tuberculous pericarditis (Strang et al., 1988). Steroids also reduce
tuberculous pleural effusions but not subsequent formation of adhesions (Lee
et al., 1988). Other uses for steroids in tuberculosis include management of
allergic drug reactions and gross enlargement of lymph nodes. Some clini
cians give steroids in genito-urinary tuberculosis with the aim of preventing
ureteric obstruction but the value of this therapy is unproven.
The use of adjunct steroid therapy in tuberculosis has been comprehen
sively reviewed by Alzeer and FitzGerald (1993).
Treatment of patients with renal or hepatic disease
It is fortunate that rifampicin, isoniazid, pyrazinamide, ethionamide and prothionamide are either metabolized or eliminated in the bile. These may there
fore be used safely at the normal doses in patients with renal impairment.
Ethambutol is mainly eliminated by the kidney but can be used in reduced
doses in patients with impaired renal function. Streptomycin and other amino
glycosides should be avoided as they are eliminated entirely by the kidney
and, in addition, are nephrotoxic in some patients. Encephalopathy is an
uncommon but serious complication of isoniazid therapy in patients with
renal failure and those on dialysis but is preventable in most, but not all, cases
by administering pyridoxine (Cheung et al., 1993).
Patients with impaired liver function may be treated with isoniazid and
ethambutol for one year, with the addition of streptomycin for the first two or
three months. The more hepatotoxic drugs rifampicin and pyrazinamide are
thus avoided, although in fact there is no evidence that these are any more
toxic in patients with impaired hepatic function. Il is important to monitor
hepatic function regularly during therapy.
The role of adjunct steroid therapy in tuberculosis
Steroids have been used in the therapy of tuberculosis for two reasons. First,
there was an assumption that, by reducing the host’s immune response, the
dormant bacilli would replicate freely and thereby become more susceptible
to killing by the drugs. There is little evidence to support this conjecture and
it has been shown that the use of steroids does not affect the outcome of mod
ern short-course chemotherapy (Fox, 1978). Secondly, steroids are used to
suppress damaging inflammatory reactions and subsequent scarring.
The use of steroids in tuberculous meningitis is controversial. Several con
trolled clinical trials have shown that they reduce mortality. They are certainly
life-saving in the presence of cerebral oedema which sometimes develops after
, the commencement of drug therapy and is probably an allergic phenomenon.
On the other hand, there is very little evidence that they reduce subsequent
Chemoprophylaxis and preventive therapy of tuberculosis
Chemoprophylaxis is defined as the prescription of antituberculosis drugs for
'uninfected people who are exposed to a risk of infection, while preventive
therapy refers to the treatment of people who have already been infected with
tubercle bacilli (as indicated by tuberculin testing) but show no clinical or
radiological evidence of active disease.
These preventive measures are principally used to protect children who are
at risk of infections, particularly those under the age of three years who are
prone to develop serious extrapulmonary forms of tuberculosis, including
meningitis.
. , , .
u
utv
The use of preventive therapy for adults, particularly those who are HIV
positive, is more controversial and is discussed in Chapter 6 (see page 132).
The role of preventive therapy in patients, other than transplant recipients
(see below), who are receiving steroids is also controversial. It is often stated
that, by compromising the immune response, steroids permit the reactivation
of latent tuberculosis but the evidence for this is weak (Bateman, 1993) and
neither short-term high-dose or long-term maintenance steroid therapy for
vitro correlates
obstructive airway disease reduced skin test reactivity or in v::ra
ccrre.z-s of
?
immunity to tuberculosis (Lowe et al., 1987).
Transplant recipients receiving steroids and other immunosuppressive
drugs are al risk of developing tuberculosis. It has been suggested that such
patients should be given isoniazid, 300 mg, and pyridoxine, 25-50 mg, daily
if they have one or more of the following: a history of inadequately treated
tuberculosis, an abnormal chest X-ray, a positive tuberculin test of more than
10 mm in diameter and recent contact with a case of active tuberculosis
(Qunibi et al., 1990).
Supportive measures in the therapy of tuberculosis
This topic is discussed in Chapter 6 (see page 128).
■ 212
Tuberculosis 213
Therapy of mycobacterial disease
Monitoring and follow-up of therapy
The purpose of monitoring drug therapy is three-fold: first, to ensure that the
patient is taking the drugs; second, to determine whether the regimen is effec
tive in the individual patient; and, third, to ensure that the drugs are having no
harmful effects. The need to monitor patients during courses of modem short
course therapy by radiology and sputum culture is questionable and, in many
regions, impossible. Likewise, the need for regular check-ups after the com
pletion of therapy for fully drug-susceptible disease has been challenged.
Even where such follow-up procedures have been applied (at considerable
expense), almost all cases of reactivation were detected when the patient pre
sented with a recurrence of symptoms (Albert et al., 1976). Accordingly,
resources should be used to ensure that every patient receives a full, super
vised course of therapy rather than for any follow-up procedures (Rouillon et
al., 1976). Patients should then be informed that although they have almost
certainly been cured they should seek medical advice promptly if symptoms
recur. On the other hand, patients with multidrug-resistant tuberculosis
require follow-up for at least two years after bacteriological cure.
Drug toxicity
Although all antituberculosis drugs have some untoward side-effects, drug
toxicity is, in general, not a serious problem and is a small price to pay for the
very real benefits of modern chemotherapy. The major side-effects arc hepato
toxicity, peripheral neuropathy, mental disturbances, rashes and fevers. Side
effects arc particularly likely to occur in HIV-positive patients.
The three principal drugs used in modern short-course regimens - iso
niazid, rifampicin, pyrazinamide - are all potentially hepatotoxic but this is
seldom a problem in clinical practice (Girling, 1989). The hepatotoxicity of
pyrazinamide was over-emphasized in the past, when larger doses were given.
It is generally recommended that liver function tests should be performed
before commencing therapy but that they need not be routinely monitored
during treatment unless the initial tests are abnormal or if the patient is an
alcoholic or has liver disease. On the other hand, drug-induced acute liver fail
ure, though rare, is extremely serious and some workers therefore advocate
monthly liver function tests during therapy (Mitchell et al., 1995).
Rifampicin may cause an influenza-like syndrome but, paradoxically, this is
less likely to occur if the drug is given daily rather than twice or thrice weekly.
Although the evidence that rifampicin is teratogenic is very limited, it is best
avoided if possible during the first three months of pregnancy. For the same
reason, women receiving rifampicin should avoid becoming pregnant. In this
respect it is important to note that this drug interferes with the action of oral
contraceptives. Alternative forms of birth control should therefore be used.
Isoniazid may cause peripheral neuritis and mild psychiatric disturbances
which are usually preventable by giving pyridoxine (vitamin B6) 10 mg daily.
Although not prescribed routinely to all patients, pyridoxine should certainly
be given to patients with liver disease, pregnant women, alcoholics, renal
dialysis patients, HIV-positive patients, the malnourished and the elderly.
The rarely used second-line drug cycloserine also causes psychiatric symp
toms, including hallucinations. Streptomycin is toxic for the eighth nerve
including that of the foetus. For this reason, streptomycin should not be given
during pregnancy.
, t . .
Ethambutol has a very important side-effect; namely, ocular toxicity.
Although this is rare if the drug is given for no more than two months at a
daily dose of 25 mg/kg body weight, or for longer at a dose not exceeding
15 mg/ko vigilance is indicated (particularly in litigation-conscious commu
nities) A code of practice, suitable for use in developed countries with the
requisite staff and facilities, has been recommended by the British Thoracic
Society:
• Pre-treatment renal function should be investigated by assay of serum
urea and/or creatinine; and ethambutol should not be given to patients
with impaired renal function.
• The recommended dose of ethambutol and duration of therapy should
never be exceeded.
. Any history of eye disease should be recorded in the notes.
• Pre-treatment visual acuity should be assessed by the Snellen test or, tor
those unable to read, by the Cambridge Low Contrast gratings. If the
patient normally wears spectacles for distant vision they should be worn
for the test. Ethambutol should not be given to patients with poor sight
who may not notice further minor deterioration of vision.
• The small risk of ocular toxicity should be explained to the patient with
an admonition to discontinue the drug if vision becomes impaired.
• A record should be made in the notes that the danger of ocular toxicity
has been explained to the patient.
.
.
• The patient’s general practitioner should be informed of the instructions
given to the patient.
• Patients complaining of visual disturbance should be referred to an
ophthalmologist.
• Routine tests of visual acuity during therapy are not recommendea.
• Ethambutol should be avoided in patients who are unsuitable for objec
tive tests of visual activity, i.e. young children and adults with language
or other communication problems.
Thiacetazone is a common cause of rashes, particularly in patients of Chinese
ethnic origin. More severe skin reactions, exfoliative dermatitis and the
Stevens-Johnson syndrome, occur in less than 0.5 per cent of patients but
there is a 10-fold increase in the incidence of these reactions in HIV-positive
patients, proving fatal in up to 3 per cent of affected cases. Thus this drug
should be avoided whenever possible and never given to a patient known to be
HIV-positive.
Drug interactions
Clinically significant interactions between the first-line antituberculosis drugs
themselves are uncommon but such reactions could well occur when more
complex regimens are used to treat multidrug-resistant tuberculosis.
Leprosy 215
214
Therapy of mycobacterial disease
Antitubcrculosis drugs may interact with drugs used to treat unrelated condi
tions. Rifampicin is the most important in this respect as it is a potent inducer
of cytochrome isoenzymes involved in lhe metabolism of many drugs. The
drugs known to be metabolized by these isoenzymes include cyclosporin, oral
contraceptives, corticosteroids, phenytoin, imidazole antifungals, theo
phylline and warfarin. The increased clearance of these drugs may lead to
therapeutic failure unless levels are adjusted, and readjusted when rifampicin
therapy ceases. Patients on oral contraceptives should be advised to use alter
native forms of birth control (see above).
For a detailed review of antituberculosis drug interactions sec Grange et al.
(1994).
L
j;
The role of drug susceptibility testing
The purpose of drug susceptibility testing, as outlined al the beginning oi this
chapter, is to determine whether the great majority of organisms in a culture
are susceptible to levels of the drugs that are achieved clinically. Tests on pre
treatment isolates will reveal whether the patient has been infected by a
resistant strain, i.e. initial or primary resistance. Tests during therapy or fol
lowing relapse will detect the emergence of resistant mutants, i.e. acquired oi
secondary resistance.
Susceptibility testing is indispensable in the assessment of new drug regi
mens and has epidemiological value in determining the pattern of resistance
in a community. The need to determine and monitor the global incidence and
distribution of drug resistance has increased recently due to the alarming
reports of multidrug resistance in several countries, for which purpose the
World Health Organization has proposed the establishment of a network of
supra-rcgional reference laboratories (Nunn and Felten, 1994). The necessity
or even desirability of testing every isolate in clinical practice is contestable.
Susceptibility testing is expensive and time-consuming and requires a high
level of technical expertise and rigid quality-control procedures. The test is
not worth performing unless high standards of accuracy can be maintained
Much harm may be done by modifying a regimen to include less effective and
more toxic drugs on the basis of a false report of resistance. Many more thera
peutic failures result from irregular medication due to poor supervision or an
erratic supply of drugs than to primary resistance (Grosset, 1978). On the
other hand, where facilities are available, and where multidrug-resistant
tuberculosis is common, such as New York, susceptibility testing of all c mical isolates is definitely indicated (Centers for Disease Control, 1993). lhe
technical procedures for susceptibility testing are described in Chapter 4.
Leprosy
Principles of chemotherapy
The aims of the chemotherapy of leprosy are similar to those of tuberculosis,
namely, the destruction of all bacilli and the prevention of relapses and the
emergence of drug resistance. As discussed in Chapter 7, the number of bacilli
in patients varies greatly and is related to the position of the patients on the
immunological spectrum. Patients Jit or near the lepromalous pole haye_aji
enormous bacillfry load and are presumed to be the most infectious.
Furthermore as they have little or n^nmninc reactivity, the baci h can onls
he destroyed by effective drug regimens. Asin the case of tuberculosis, some
bacilli persist for long periods of time, even in lhe presence of the bactencida
drUThefead?esreffe^ drug for leprosy was chaulmoogra oil, der^ fr^
the fruit of Hydnocarpus wightiana. This drug, the origin of which has
inspired various legends, is of some efficacy in tuberculoid leprosy and may
actPby stimulating the immune responses. In 1943 Faget and
introduced promin which, although effect.ve, was rather toxic: and had to
given by intravenous injection. The same workers subsequently isolated the
Active principle of promin - diaminodiphenyLsulphone (DDS,
which was cheap, very effective,virtually non-tox.c and c^WbeU^en by
mouth In view oFtHese properties, dapsonc became the mainstay of
antileprosy therapy, even though it is bacteriostatic rather than bactericidal
Side-effects are uncommon with dapsone, although cases of haemolytic
anaemia and other blood disorders, hepatitis, dermatitis and Pjycho^,s
sionally occur. Agranulocytosis is a rare but serious complication. The term
‘dapsone syndrome’ is applied to a skin rash and fever, usually occurrin0 2 8
weeks afte? starting therapy and sometimes accompanied by lymphadenopa
thy, hepatomegaly, jaundice and/or mononucleosis.
hmnrene) and
The two other first-line drugs are clofazimine (B663, Jamprene) ana
rifampicin. The former is weakly bactericidal; antibactenal
leprae is only demonstrable in human beings after 50 days of therapy.
Clofazimine-resistant leprosy, though reported, is rare. It has a veiy long half Ide,
70 days, and is eliminated in the urine and faeces. It is also an;,:*nflamin^^
suppresses erythema nodosum leprosum reactions (see page 154) 11 ^asionafiy
causes gastrointestinal disturbances, relieved to some extent by lakmg the drug
with a meal or glass of milk. It has the more serious’ d,sad^[Xfi^
discolouration which many patients find unacceplabie,
cornea urine, sweat and tears also occurs. Infants bom to mothers receivi g
c““iTiSXtricyidp&
comitant mpid reduction in infectivity. Indeed, about 99J MceriW
are killed within a week of administering a single dose of 600-1500 mg. of
nfamnicin Unfortunately a minority of bacilli ‘persist’ and small numbers of
not ciprofloxacin), clarithromycin and minocycline These ary^0™^
for the treatment of patients who refuse to lake dofazmnne or foi■ who
216
Therapy of the inycobacterioses
Therapy of mycobacterial disease
hepatic toxicity and gastric irritation, and the availability of other drugs, their
use is no longer recommended (World Health Organization, 1994).
Table 10.5 WHO recommended chemotherapeutic regimens for leprosy (WHO, 1994)
Standard regimens
Paucibacillary leprosy
Drug regimens for leprosy
Until around 1976, dapsone monotherapy was the standard treatment for all
forms of leprosy. The emergence of dapsone resistance in multibacillary
patients had been reported and, although the incidence appeared to be very
low, the World Health Organization Expert Committee on Leprosy (1977)
stressed the need for studies on multidrug therapy for proven dapsone
resistant cases. At that time there were disturbing reports from some regions
suggesting that the incidence of such resistance was higher than previously
suspected and that it was increasing. The World Health Organization therefore
sponsored a number of studies which confirmed the seriousness of the
problem (World Health Organization Scientific Working Group on the
Chemotherapy of Leprosy, 1982). Such emergent, DLsecondaryJJnig_res_istance is confined to patients^with multibacillary cases, as only_they havelarge
enough numbers of replicating bacilli to make it likely that a mutation will
.occur. On the other hand, when such bacilli are transmitted to other persons,
any form of leprosy may develop. Such primary resistance, unfortunately,
now occurs in several countries (Waters et al., (978).
In view of the increasing frequency of both primary and secondary dapsone
resistance, multidrug therapy (MDT) is now considered essential for the treat
ment of all forms of leprosy. Different regimens for multibacillary and pauci
bacillary cases have been recommended by the World Health Organization
(1982, 1994), and are now used routinely in most countries where leprosy
occurs. These MDT regimens arc summarized in Table 10.5.
For the purposes of deciding which multidrug regimen to use, patients with
negative slit-skin smears are classified as paucibacillary (PB) while those
positive at any site are classified as multibacillary (MB). Relapsing patients
with previous MB disease treated with dapsone alone are classified as MB lor
the purpose of retreatment, even if all slit-skin smears are negative. In prac
tice, many patients are classified as PB or MB on the basis of clinical features
rather than on microscopical findings.
Patients with MB disease require therapy for two years. As there is a huge
bacillary load, there is a high chance that mutation to rifampicin resistance
will develop. It is therefore essential to use three drugs. Patients with PB dis
ease are treatable by short course regimens as any persisters are almost cer
tainly destroyed by the patients’ immune defences. As the bacillary load is
small, the chance of drug resistance developing by mutation is very low.
Accordingly a third drug is not essentia).
As in the case of tuberculosis, relapses occasionally occur, particularly in
MB cases (Waters, 1995) but, when tested for drug susceptibility, the bacilli
from such cases are usually fully susceptible. In addition, residual antigen
may continue to induce nerve damage and leprosy reactions after therapy is
complete (Shetty et al., 1992).
Non-compliance is a recurring problem and thus there is still a need for
shorter or intermittent regimens. The advantages and problems of standard
217
Multibacillary leprosy
Alternative regimens
Rifampicin resistance
or toxicity
Refusal to take
clofazimine
Rifampicin
Dapsone
- 600 mg monthly, supervised
- 100 mg daily, unsupervised
Rifampicin
Dapsone
Clofazimine
- 600 mg monthly, supervised
- 100 mg daily, unsupervised
- 300 mg monthly, supervised AND
50 mg daily, unsupervised
For 6 months, daily, under supervision:
- 50 mg, and TWO of
Clofazimine
- 400 mg
Ofloxacin
-100 mg
Minocycline
Clarithromycin - 500 mg
Foran additional 18 months, daily, under
supervision:
- 50 mg and ONE of
Clofazimine
- 400 mg
Ofloxacin
- 100 mg
Minocycline
Replace clofazimine in either standard
regimen with daily, supervised:
Ofloxacin
- 400 mg OR
Minocycline
-100 mg
multidrug therapy and the possibilities for improved therapy based on altemalive drugs such as ofloxacin, minocycline and clarithromycin are discussed by i
and Grosset (1990), Grosset (1994) and the World Health Organization (1994).
Chemoprophylaxis of leprosy
Household contacts of leprosy patients should be examined tor signs of the dis
ease and if none are found, they should be advised to seek medical attention if
any signs develop. Dapsone has been given prophylactically with some success
but in view of the medical and administrative problems generated by prophy
laxis programmes, as well as the problem of dapsone resistance this control
measure is no longer recommended (World Health Organization, 1994).
Drug therapy of leprosy reactions
This subject is discussed in Chapter 7 (see page 154).
Therapy of the mycobacterioses
The treatment of the mycobacterioses, or opportunist mycobacterial diseases,
is complicated by the problems of diagnosis and the difficulty m distinguish
ing between true disease and colonization; by the frequent occurrence of
218
Therapy of the mycobacterioses 219
Therapy of mycobacterial disease
serious underlying immune defects or other predisposing conditions; by the
natural resistance of many mycobacteria to the available antimicrobial agents;
the variation of this natural resistance within species and by the lack of corre
lation between in vitro resistance and response to therapy in vivo. The prob
lem has been compounded by the frequent occurrence of such disease in
patients with acquired immune deficiency syndrome (AIDS).
From the point of view of therapy, the mycobacterioses are divisible into
three main groups:
1. pulmonary (and occasionally solitary non-pulmonary) disease due to
slowly growing mycobacteria in HIV-negative patients;
2. disease, usually disseminated and due to the M. avium complex, in HIV
positive patients;
3. disease due to the rapidly growing species M. chelonae and M. fortuitum.
Treatment of the two named diseases, Buruli ulcer and swimming-pool granu
loma, is discussed in Chapter 9.
Not all mycobacterial lesions require chemotherapy. Localized lympha
denitis in children almost always responds to surgical excision which is usu
ally undertaken for diagnostic purposes. It is important that the whole
infected node is removed as simple incision and drainage often leads to
sinus formation, scarring and delayed healing. Children should be followed
up for a few years to ensure that the disease does not recur or spread.
Localized post-traumatic or post-injection lesions also usually occur in other
wise healthy individuals and tend to resolve spontaneously. Nevertheless,
healing may be prolonged and complicated by the presence of discharging
sinuses. Small lesions may be cured by simple excision and larger ones by
curettage.
Pulmonary disease due to slowly growing species
The usual organisms are the M. avium complex, M. kansasii, M. xenopi and
M. malmoense. In view of in vitro resistance of the M. avium complex to most
of the antituberculosis agents, it was considered that standard triple anti
tuberculosis therapy would be of no value. Instead, regimens containing five
or six empirically chosen drugs were administered for up to 3 years
(Davidson, 1976). This regimen was undoubtedly successful in a high pro
portion of patients who completed the therapy. Unfortunately many patients
developed adverse drug reactions and compliance was poor. It was subse
quently found that a more acceptable triple regimen of rifampicin, isoniazid
and ethambutol is as effective for uncomplicated cases, provided that therapy
is continued with all three drugs for at least 18 months (Engbaek et al., 1984).
This regimen has also proved effective for the treatment of pulmonary disease
due to other slowly growing species, including M. kansasii (Banks et al.,
1983), M. xenopi (Smith and Citron, 1983) and M. malmoense (Banks et al.,
• 1985).
The inclusion of ethambutol for the whole duration of therapy is essential
,
.
,
.
,
,
•
r>
•f
•••..« I..
bacterial cell. The role of isoniazid is less certain: disease due to M. kansasii
infection responds well to a 9-month regimen of rifampicin and ethambutol
(British Thoracic Society, 1994).
.
.
Despite the efficacy of therapy, relapse is not uncommon, especially in the
case of M. xenopi. Thus surgical intervention, though rarely resorted to nowa
days in the case of uncomplicated pulmonary tuberculosis, should be seri
ously considered for localized pulmonary lesions due to other mycobacteria
(Moran et al., 1983).
..................
.
f
Experience with other drugs and species is limited and anecdotal. Strains ot
M. kansasii, M. xenopi, M. malmoense, the M. avium complex and M. scrofulaceum are usually susceptible to clarithromycin and fluroquinolones in vitro
and there is anecdotal evidence of their efficacy in vivo. Co-trimoxazole has
also been used in the successful treatment of cases of disease due to M. xenopi
(Grange, 1984).
For further details on the therapy of disease due to the slowly growing
species see Heifets (1994) and Banks (1994).
Rapidly growing mycobacteria
There have been very few controlled clinical trials ol therapeutic regimens tor
disease due to the rapidly growing mycobacteria M. chelonae and M.fortuitum. Most cases are treated on the basis of anecdotal experience or by trial and
error As the drugs are almost always used in combination, it is difficult to
assess the contribution of each agent to the result. Furthermore, the optimum
duration of therapy has not been established and must therefore be based on
the clinical response in individual patients. Regimens containing antitubercu
losis drugs, as used for treating disease due to slow growing opportunistic
pathogens, are ineffective against the rapid growers.
Erythromycin has been used successfully together with amikacin or
aentamicin (Wolinsky, 1979) or with co-trimoxazole (Azadian et al., 1981;
Jackson et al., 1981). It has been suggested that the sulphonamide component
of co-trimoxazole acts by enhancing the ability of the phagocytic cells to kill
the bacteria (Tice and Solomon, 1979). It therefore appears to be of benefit
even if the organism is resistant to it in vitro (Jackson et al., 1981).
Doxycycline has been used with amikacin to treat infections caused by
M. chelonae and M. fortuitum (Dalovisio et al., 1981) and it was used with
erythromycin and co-trimoxazole in the successful treatment of a dissemi
nated M. chelonae infection in a renal transplant recipient (Azadian et al.,
Some beta-lactam antibiotics are active against the rapidly growing
pathogens. Of the many cephalosporins, cefoxitin and cefmetazole are the
most active in vitro and the former was used successfully with amikacin in the
treatment of sternal infections due to M. chelonae following cardiac surgery
(Kuritsky et al., 1983). Imipenem is active against M. chelonae and the fluro
quinolones against M. fortuitum (Yew et al., 1990) but clinical experience
with these drugs is limited. For further details ol drug susceptibilities of M.
chelonae and M. fortuitum see Cynamon and Klemens (1991) and for a
C
K-fr-'C'nrlnnd nrvl 1611 ri t'/L'C C ( 1 Q9 3J
. 222
Therapy of mycobacterial disease
Fox, W. 1978: The current status of short-course chemotherapy. Bulletin of the
International Union Against Tuberculosis 53, 268-80.
Fox, W. 1990: Drug combinations and the bioavailability of rifampicin. Tubercle 71,
241-5.
Freerksen, E. and Rosenfeld, M. 1977: Leprosy eradication project of Malta: first pub
lished report after 5 years running. Chemotherapy 23, 356-86.
Girling, D.J. 1989: The chemotherapy of tuberculosis. In Ratledge, C., Stanford, J.L.
and Grange, J.M. (eds.), The biology of the mycobacteria. Vol. 3. New York:
Academic Press. 285-323.
Goble, M., Iseman, M.D., Madsen, L.A. et al. 1993: Treatment of 1712 patients with
pulmonary tuberculosis resistant to isoniazid and rifampicin. New England Journal
of Medicine 328, 527-32.
Grange, J.M. 1984: Therapy of infections caused by ‘atypical’ mycobacteria. Journal
of Antimicrobial Chemotherapy 13, 308-10.
Grange, J.M., Winstanley, P.A. and Davies, P.D.O. 1994: Clinically significant drug
interactions with antituberculosis agents. Drug Safety 11, 242-51.
Grosset, J.H. 1978: Should a sensitivity test be made for each new case of tubercu
losis? Bulletin of the International Union Against Tuberculosis 53, 200-1.
Grosset, J.H. 1994: Progress in the chemotherapy of leprosy. International Journal of
Leprosy 62, 268-77.
Heifets, L.B. 1994: Antimycobacterial drugs. Seminars in Respiratory Infections 9,
84-103.
Iseman, M.D., Corpe, R.F., O’Brien, R.J. et al. 1985: Disease due to Mycobacterium
avium-intracellulare. Chest 87 (Suppl. 2), 139s-49s.
Jackson, P.G., Keen, H., Noble, C.J. and Simmons, N.A. 1981: Injection abscesses due
to Mycobacterium chelonei occurring in a diabetic patient. Tubercle 62, 277-9.
Ji, B. and Grosset, J.H. 1990: Recent advances in the chemotherapy of leprosy.
Leprosy Review 61, 313-29.
Kumarvelu, S., Prasad, K., Khosla, A. et al. 1994: Randomised controlled trial of
dexamethasone in tuberculous meningitis. Tubercle and Lung Disease 75, 203-7.
Kuritsky, J.N., Bullen, M.G., Broome, C.V. et al. 1983: Sternal wound infection due
to organism of the Mycobacterium fortuitum complex. Annals of Internal Medicine
98, 938-9.
Lee, C.-H., Wang, W.-J., Lan, R.-S. et al. 1988: Corticosteroids in the treatment of
tuberculous pleurisy: a double-blind, placebo-controlled, randomised study. Chest
94, 1256-9.
Lowe, J.G., Beck, J.S., Gibbs, J.H. et al. 1987: Skin test responsiveness to four new
tuberculins in patients with chronic obstructive airway disease receiving short term
high doses of, or long term maintenance treatment with, prednisolone: clinical
appearance and histometric studies. Journal of Clinical Pathology 40, 42-9.
Masur, H., and the US Public Health Service Task Force on Prophylaxis and Therapy
for M. avium Complex. 1993: Recommendations on prophylaxis and therapy for
disseminated M. avium complex disease in patients infected with the HIV (Special
Report). New England Journal of Medicine 329, 898-904.
McFarland, E.J. and Kuritzkes, D.R. 1993: Clinical features and treatment of infection
due to Mycobacterium fortuitum/chelonae complex. Current Clinical Topics in
Infectious Diseases 13, 188-202.
Mitchell, I., Wendon, J., Fitt, S. and Williams, R. 1995: Anti-tuberculous therapy and
acute liver failure. Lancet 345, 555-6.
Mitchison, D.A. 1985: Hypothesis: the action of antituberculosis drugs in short course
chemotherapy. Tubercle 66, 219-25.
Moran, J.E, Alexander, L.G., Staub, E.W. et al. 1983: Long-term results of pulmonary
resection for atypical mycobacterial disease. Annals of Thoracic Surgery 35,
597-604.
References
223
Morse, D.L. 1994: Multidrug resistance. The New York experience. In Porter, J.D.H.
and McAdam, K.P.W.J. (eds.). Tuberculosis: back to the future. Chichester: John
Wiley. 225-30.
Nunn, P. and Felten, M. 1994: Surveillance of resistance to antituberculosis drugs in
developing countries. Tubercle and Lung Disease 75, 163-7.
O’Brien, R.J. 1994: Drug-resistant tuberculosis: etiology, management and preven
tion. Seminars in Respiratory Infections 9, 104-12.
Onyebujoh, P.C., Abdulmumini, T., Robinson, S. et al. 1995: Immunotherapy with
Mycobacterium vaccae as an addition to chemotherapy for the treatment of pul
monary tuberculosis under difficult conditions in Africa. Respiratory Medicine 89,
199-207.
Pattyn, S.R., Dockx, P., Rollier, M.T. etal. 1976: Mycobacterium leprae persisters after
treatment with dapsone and rifampicin. International Journal of Leprosy 44, 154-8.
Qunibi, W.Y., Al-Sibai, M.B.,Taher, S. etal. 1990: Mycobacterial infection after renal
transplantation - a report of 14 cases and a review of the literature. Quarterly
Journal of Medicine 77,1039-60.
Rouillon, A., Perdrizet, S. and Parrot, R. 1976: Transmission of tubercle bacilli: the
effects of chemotherapy. Tubercle 57, 275-99.
Shetty, V.P., Suchitra, K„ Uplekar, M.W. and Antia, N.H. 1992: Persistence of
Mycobacterium leprae in the peripheral nerve as compared to the skin of multidrugtreated leprosy patients. Leprosy Review 63, 329-36.
Smith, M.J. and Citron, K.M. 1983: Clinical reviews of pulmonary disease caused by
Mycobacterium xenopi. Thorax 38, 373-7.
Sphalinger, H. 1922: Note on the treatment of tuberculosis. Lancet 1, 5-8.
Stanford, J.L., Onyebujoh, P.C., Rook, G.A.W. et al. 1993: Old plague, new plague
and a treatment for both. AIDS 7, 1275-7.
Stanford, J.L., Stanford, C.A., Rook, G.A.W. and Grange, J.M. 1994: Immunotherapy
for tuberculosis. Investigative and practical aspects. Clinical Immunotherapeutics
1 430-40
Strang, J.G., Kakaza, H.H.S., Gibson, D.G. et al. 1988: Controlled clinical trial of
complete open surgical drainage and prednisolone in the treatment of tuberculous
pericardial effusion in Transkei. Lancet ii, 1418-22.
.
Tice, A.D. and Solomon, R.J. 1979: Disseminated Mycobacterium chelonei infection:
response to sulphonamides. American Review of Respiratory Disease 120,
197-201.
...
Uplekar, M.W. and Shepard, D.S. 1991: Treatment of tuberculosis by private general
practitioners in India. Tubercle 72, 284-90.
Waters, M.F.R. 1995: Relapse following various types of multidrug therapy in multibacillary leprosy. Leprosy Review 66, 1-9.
Waters, M.F.R., Rees, R.J.W., Pearson, J.M.H. et al. 1978: Rifampicin for lepromatous leprosy: nine years’ experience. British Medical Journal 1, 133-6.
Winder, EG. 1982: Mode of action of the antimycobacterial agents and associated
aspects of the molecular biology of the mycobacteria. In Ratledge, C. and Stanford,
J.L. (eds.). The biology of the mycobacteria. Vol. 1. London: Academic Press.
353^38.
.
Wolinsky, E. 1979: State of the art: non-tuberculous mycobacteria and associated
diseases. American Review of Respiratory’ Disease 119, 107-59.
World Health Organization Expert Committee on Leprosy 1977: Fifth Report.
Technical Report Series No. 607. Geneva: World Health Organization.
World Health Organization Scientific Working Group on the Chemotherapy ot
Leprosy. 1982: Chemotherapy of leprosy for control programmes. Technical Report
Series No. 675. Geneva: World Health Organization.
World Health Organization. 1991: WHO model prescribing information. Drugs used
in
hnrterial diseases. Geneva: World Health Organization.
220
Disseminated disease
>•
f
References
Therapy of mycobacterial disease
The majority of cases are caused by the M. avium complex in AIDS patients.
The design of suitable chemotherapeutic regimens has been hampered by the
difficulty in assessing the drug susceptibility of members of this complex. A
number of regimens based on rifabutin and the antileprosy drug clofazimine
had been described (Iseman et al., 1985). Subsequently, the macrolides clari
thromycin and azithromycin were found to be highly effective, and should be
included in all regimens (Masur and the US Public Health Service Task Force
on Prophylaxis and Therapy for M. avium Complex, 1993). Companion
drugs, essential to prevent emergence of drug resistance should, if possible, be
selected on the basis of in vitro susceptibility tests (Heifets, 1994).
Disseminated disease due to other species in HIV-positive patients and such
disease in HIV-negative patients is uncommon and treatment is based on the
results of in vitro drug susceptibility tests.
I,
Immunotherapy of mycobacterial disease
In view of the problems of non-compliance with therapy and the emergence ol
drug resistance in many countries, alternative forms^of therapy for tubercu
losis are urgently needed. The most promising is immunotherapy. The
immunological basis of such therapy and the evidence that the jiqupajthogenic rapidly growing species M. vaccae has the properties required of
anTmmunotherapeutic agent are summarized in Chapter 5.
This form of immunotherapy is a useful adjunct to chemotherapy of fully drugsusceptible tuberculosis but more impressive results have been obtained in the
treatment of drug-resistant disease. Experience in Iran showed that immuno
therapy gave good results in multidrug-resistant tuberculosis of recent onset but
was less effective in very chronic cases with gross pulmonary scarring.
Nevertheless, a 40 per cent cure rate was obtained, compared with a 1 per cent
cure rate previously obtained with the limited range of available drugs (Etemadi
et al., 1992). Immunotherapy is particularly effective in regions where
chemotherapy is very suboptimal. In Kano, Nigeria, where most of the available
drugs were time-expired or fake, the mortality was very substantially reduced
(Onyebujoh et al., 1995): 19 of 47 (40 per cent) who received chemotherapy
alone died during the ensuing 10-14 months while none of 34 who received both
chemotherapy and immunotherapy died. A study in the same region showed that
immunotherapy had a very beneficial effect in HIV-related tuberculosis: all 8
HIV-positive patients who were given the immunotherapy were alive after 2
years while 8 of 9 patients receiving placebo were dead and the other patient
could not be traced (Stanford et al., 1993). For more details on the theoretical
and practical aspects of immunotherapy see Stanford et al. (1994).
221
Alzeer. A.H. and FitzGerald, J.M. 1993: Corticosteroids and tuberculosis: risks and
use as adjunct therapy. Tubercle and Lung Disease 14. 6-11.
Azadian B.S., Beck, A., Curtis, J.R. et al. 1981: Disseminated infection with
Mycobacterium chelonei in a haemodialysis patient. Tubercle 62.281-jk
Banks, J. 1994: Environmental mycobacteria. In Davies, P.D.O. (ed.). Clinical
tuberculosis. London: Chapman & Hall, 265-75.
.
Banks J , Hunter, A.M., Campbell, I.A. et al. 1983: Pulmonary infection with
Mycobacterium kansasii in Wales 1970-9: review of treatment and response.
Banks™P.A. and Smith. A.P. 1985: Pulmonary infection with
Mycobacterium malmoense - a review ot treatment and response. Tubercle 66,
197—203.
Bartmann, K. (ed.) 1988: Antituberculosis drugs (Handbook of experimental pharma
cology,vol. 84). Berlin: Springer Verlag.
.
.
Bateman E.D. 1993: Is tuberculosis prophylaxis necessary tor patients receiving
corticosteroids for respiratory disease? Respiratory' Medicine 87, 485-7.
British Medical Research Council Working Party on Tuberculosis of the Spine. 1978.
Seventh Report. A controlled trial of anterior spinal fusion and debridement in the
surgical management of tuberculosis of the spine in Patient$ °" standard
chemotherapy: a study in two centres in South Atnca. Tubercle 59, 79 *B->British Thoracic Society. 1994: Mycobacterium kansasii pulmonary infection: a
prospective study of the results of nine months of treatment with rifampicin and
ethambutol. Thorax 49,442-5.
. .
Centers for Disease Control. 1993: Initial therapy for tuberculosis in the era of multi
drug resistance. Recommendations of the Advisory Council tor the Elimination o
Tuberculosis. Morbidity and Mortality Weekly Report 42(RR-7), 1-8.
Cheung. W.C., Lo, C.Y., Lo, W.K. et al. 1993: Isoniazid induced encephalopathy in
dialysis patients. Tubercle and Lung Disease 74, 136-9.
Cynamon, M.H. and Klcmcns, S.P. 1991: Drug susceptibility tests for Mycobacterium
fortuitum and Mycobacterium chelonae. In Heifets, L.B. (cd.), Drug susceptibility
in the chemotherapy of mycobacterial infections. Boca Raton: CRC Press.
Dalovisio, J.R., Pankey, G.A., Wallace, R.J. and Jones, D.R. 1981: Clinical useful
ness of amikacin and doxycycline in the treatment of infection due to
Mycobacterium fortuitum and Mycobacterium chelonei. Review of Infectious
Diseases 3, 1068-74.
David, H.L. 1970: Probability distribution ot drug-resistant mutants in unselectcd
populations of Mycobacterium tuberculosis. Applied Microbiology 20, 810-4.
Davidson, P.T. 1976: Treatment and long-term follow-up of patients with atypical
mycobacterial infections. Bulletin of the International Union Against Tuberculosis
Davte^P.D.O., Girling, D.J. and Grange, J.M. 1995: Tuberculosis. In Weatherall,
References
D.J., Ledingham, J.G.G. and Warrel, D.A. (eds.), Oxford textbook of medicine. 3rd
cdn. Oxford: Oxford University Press. 638-61.
Dooley S.W. and Simone, P.M. 1994: The extent and management of drug-resistant
tuberculosis: the American experience. In Davies, P.D.O. (ed.). Clinical tubercu
losis. London, Chapman & Hall. 171-89.
Engbaek, H.C., Vergmann, B. and Bentzon, M.W. 1984: Lung disease caused by
Mycobacterium avium/Mycobacterium intracellulare. European Journal oj
EsS™A7fXTMl4D\las. A. and Medina. B 1975: Mortality front tuber-
Albert, R.K., Iseman, M., Sbarbaro, J.A. et al. 1976: Monitoring patients with
tuberculosis for failure during and after treatment. American Review of Respiratory
culous meningitis reduced by steroid therapy. Pediatrics 56, 1050-5.
Etemadi, A.. Farid, R. and Stanford, J.L. 1992: Immunotherapy for drug-resistant
..■U.^^ulncic
I nnrft 340
1360—1
224
Therapy of mycobacterial disease
World Health Organization. 1994: Chemotherapy of leprosy. Technical Report Series
No. 847. Geneva: World Health Organization.
Yew, W.W., Kwan, S.Y.L., Wong, P.C. and Lee, J. 1990: Ofloxacin and imipenem in
the treatment of Mycobacterium fortuitum and Mycobacterium chelonae lung infec
tions. Tubercle 71, 131-3.
r
u
"f
it!'
ii»
■i
•It
1“
fin
u
"t
:i:
|H
.'i
I
Index
A60 antigen 14
Abdominal tuberculosis 180-1
Abscesses: post-injection 195;
spinal tuberculosis 172
Absolute concentration method,
drug susceptibility testing 72
Acid-fastness 11,24
Acid phosphatase activity 70
(Table)
Acquired drug resistance 124
Acquired immunodeficiency
syndrome see HIV infection
Actively secreted proteins 15
Acute-phase reactants, tuberculosis
160
Acylated trehaloses 20-1
Addison’s disease 183
Adherence, to therapy 13O-I
Adjunct therapy, steroids 210
Adjuvants 17, 85, 105
Advisory Council for Elimination
of Tuberculosis 209
Affinity chromatography 14
Africa: HIV infection and
tuberculosis 122-3
Agalactosyl immunoglobulin G
104
AIDS see HIV infection
Air-water interfaces 22
Alcoholics, peritoneal tuberculosis
181
Allantoinase activity 71 (Table)
a-L-fucosidase activity:
Mycobacterium marinum 45-6,
69, 70 (Table); Mycobacterium
szulgae 47, 70 (Table)
a-mycolic acids 19
Amides 26
Amikacin 219
Amino acids 26
Amplification techniques, nucleic
acids 29
Anaesthesia, leprosy 140, 155
Anergy 96: Buruli ulcer 103;
leprosy 100; tuberculosis 102
Animal inoculation 67
Annual incidence, tuberculosis
116, 117
Annual infection rate, tuberculosis
116-17, 118
Anonymous mycobacteria see
Environmental mycobacteria;
Opportunist mycobacterial
infections
Antibodies 104-5
Anticoagulant therapy 214
Antigens 12-14, 41:
Mycobacterium leprae 55-6;
protective 95-6; recognition
79-81
Anti-inflammatory drugs, leprosy
154-5
Antimycobacterial drugs 204-20
see also Chemotherapy and
named drugs:
multidrug-resistant tuberculosis
209-10
Apical tuberculosis 164
Approved lists of bacterial names
12,41-2
Arabinogalactan 17
Armadillos 126, 138, 139;
lepromins from 153
Arthritis, mycobacterioses 195
Artificial pneumothorax 166
Arylsulphatase activity 70 (Table),
71 (Table)
Asian immigrants, tuberculosis
168-9
Asian Mycobacterium tuberculosis
43
Asparagine 26
Aspergilloma. Mycobacterium
xenopi 197
Attenuation indicator lipid 21, 25
Auramine staining 65
Autoimmune diseases 106, 201
Autonomic nerve damage, leprosy
152
Azithromycin: multidrug-resistant
tuberculosis 210;
mycobacterioses 220
Bacille Calmette-Guirin 44:
historical aspects 5
Bacille Calmette-GuSrin
vaccination 87, 97-8: efficacy
88-9,93; vs environmental
mycobacteria 200; leprosy 95-6,
104, 132; vs mycobacterial
lymphadenitis 196; osteitis 176;
tuberculin reaction 160
Bactec 460/TB 67
Bacterial index, leprosy 152
Bacteriophages 31-2, 33 (Fig.)
Base pairs, mycobacteria 28
Battey bacillus 47
Bazin's disease (erythema
induratum) 102, 182
B-cells 79
Beg gene 106
Beading. Mycobacterium kansasii
Bell’s palsy, leprosy 145 (Fig.)
Benzalkonium chloride 66
Beta-lactam antibiotics 219
Biblical leprosy 138, 182
Biofilms 22
Biopsies 63: environmental
mycobacteria 190; leprosy 151;
microscopy 65; peritoneum 181;
pulmonary mycobacterioses 196
Bites, leprosy 140
Blindness, leprosy 140-1, 155
Blood specimens 64
Bone marrow specimens 64
Bones: mycobacterioses 197;
tuberculosis 172-6
Borderline lepromatous leprosy 98.
146
Borderline tuberculoid leprosy 98.
144-5
Bovine tuberculosis see also
Milk-bome tuberculosis;
Mycobacterium bovis:
epidemiology 125—6; Royal
Commission on 4
Breast, leprosy 147
British Thoracic Society, code of
practice on ethambutol 213
Bronchial hygiene 197
Bronchopneumonia, tuberculosis
163
Bronchoscopy 63, 160-1. 196
Bruce, Robert (King of Scotland)
139 (Fig.)
Burulin 191
Buruli ulcer 190-3: distribution
189; immune spectrum 103
Butcher’s warts (prosector’s warts)
161, 182
Cachectin (TNF-tx) 81-2. 89-91,
109
Calcification, tuberculosis 162
Capillary blood flow, leprosy 152
Cardiac surgery, mycobacterioses
195
Cardiac tuberculosis 183
Carotenoids 23
Case-finding 129
Cattle, tuberculosis 125
Cavities, tuberculosis 87. 165-6
Cell lysis 78.96, 109
Cell-mediated immune responses
78-84
Cell-mediated immunity and
delayed type hypersensitivity
87-9
<^^11
rww/'oK'ir'fr’fs n 1^—1 A
Preface to the First Edition
Preface to the Second
Edition
The genus Mycobacterium, despite great advances in medical science, con
tinues to be a major cause of misery and suffering throughout the world.
Leprosy and tuberculosis attack the human race with undiminished vigour
while several other species of mycobacteria are emerging as important causes
of life-threatening disease.
It was originally envisaged that this book should be a revised version of my
previous work Mycobacterial Diseases, published by Edward Arnold in 1988.
In practice, there have been so many major developments in the subject over
the last eight years that the text has been almost entirely rewritten. In particu
lar, the drug treatment of leprosy, tuberculosis and other mycobacterial dis
ease has become much more rational and considerable-advances-have been
made in immunology, facilitated by the introduction of monoclonal anti
bodies, cell cloning techniques and modern ‘genetic engineering’procedures.
In addition, there have been advances in the ecology, biochemistry, epidemi
ology and classification of the mycobacteria. Sadly, though, there is an
increasing gap between the ‘high-tech’ researchers and those responsible for
the basic care of the victims of mycobacterial disease. I hope this book will
help to bridge that gap.
This monograph provides a review of the mycobacteria themselves, their
place in the environment, the way in which they interact with the living host,
the nature of the diseases they cause and the available means of diagnosing,
preventing and curing such disease. It is intended for both undergraduate and
postgraduate students seeking a general account of the mycobacteria and the
diseases they cause, for the clinician wishing to understand the underlying
mechanisms of the pathogenesis of the diseases, for the epidemiologist and
health care administrator wishing to appreciate the nature and magnitude of
the public health problems posed by the diseases, and for the microbiologist
providing a clinical service. The potential researcher will find an account of
the exciting developments in the science of mycobacteriology and, more
importantly, will become aware of the many gaps in our present-day
knowledge!
In 1988, when the First Edition of this book was published, few would have
predicted the enormous surge in interest in mycobacterial diseases that started
in the early 1990s and culminated in the declaration of tuberculosis as a global
emergency by the World Health Organization in 1993. A principal reason for
the current interest and concern is the profound impact that the HIV/AIDS
pandemic has had on the incidence and nature of mycobacterial disease in
both the developing and the developed world. It is clear that, unless radical
new control measures are introduced, HIV-related tuberculosis will have a
catastrophic effect on human health worldwide in the twenty-first century. As
a result, many researchers are attempting to develop new diagnostje^prevenlive and therapeutic measures by the application oi molecular technology.
Others are concerned with the infrastructure of health care and of making the
best use of available tools for disease control. Interest is not limited to
tuberculosis; as a result of the HIV/AIDS pandemic, other mycobacterial dis
eases that were once regarded as little more than curiosities are also now of
major and increasing importance.
The one mycobacterial disease that has declined in incidence is leprosy,
thanks to the widespread use of multidrug therapy. Nevertheless, there is still
a pressing need to continue with the provision of care for those with this dis
ease, many of whom will suffer from its effects for decades to come even
though they are bacteriologically cured. Tuberculosis has taught us the danger
of losing interest in an infectious disease before it, and its consequences, are
truly eradicated.
As a result of this upsurge of concern, many scientists and medical practi
tioners are currently taking a serious interest in the mycobacteria and the dis
eases that they cause. The purpose of this book, like that of the First Edition,
is to serve as a general introduction to this growing discipline. I hope it will
prove invaluable to those about to embark on journeys of basic scientific dis
covery as well as to those attempting to control the mycobacterial diseases,
often under very difficult conditions.
John M. Grange
London, 1988
John M. Grange
London, 1996
- Media
RF-TB-1.21.pdf
Position: 6150 (1 views)