New views on the pathogenesis of acute otitis media

and its complications

Martine Franqois
Service ORL, H6pital R o b e r t Debrk, Universiti: Paris VII, Paris, France

The pathogenesis of acute otitis media is complex and multifactorial. Bacteria infecting the middle ear come from the
nasopharynx via the eustachian tube. This colonization is facilitated by bacterial adherence on the pharyngeal and
the eustachian tube cells. Otitis media is characterized by inflammation of the middle ear with an infiltration of the
subperiosteal space by leukocytes, macrophages and mast cells. The effusion contains great amounts of inflammatory
mediators (eicosanoids, cytokines, histamine). Elimination of the effusion and/or the bacteria is based on non-specific
factors such as mucociliary clearance and phagocytosis, and on a specific immune response which apparently is not the
same for Haemophilus influenzae and for Streptococcus pneumoniae. The main complications of acute otitis media are
otitis media with effusion, mastoiditis, sensorineural hearing loss and meningitis.
Key words: Acute otitis media, pathogenesis, adherence, mucociliary clearance, immunoglobulin, inflammation

INTRODUCTION
Acute otitis media (AOM) is the most common organic
disease in infants and children. During the past decade,
many important contributions have been made in
the area of its pathogenesis, which is complex and
multifactorial. Studies have been focused mainly on the
bacterial colonization of the middle ear and on the
middle ear reaction to this colonization. The under-
standing of the pathogenesis of middle ear infection will
lead to a better therapy for this disease and better
prevention of both complications and recurrences.
BACTERIAL COLONIZATION OF THE MIDDLE EAR
All experimental and clinical data are leading towards
the conclusion that bacteria responsible for AOM come
from the nasopharynx via the eustachian tube.
Corresponding author and reprint requests:
Dr Martine Francois, Service ORL, HBpital Robert Debre,
48 boulevard Seruier, 75 019 Paris, France
Tel: +33 1 40 03 24 49 Fax: +33 1 40 03 47 17
Bacteria responsible for AOM are the same as those in the
nasopharynx
Nasopharyngeal colonization with potential middle ear
pathogens appears to be a prerequisite for their ability
to infect the middle ear [l]. In episodes of AOM,
the bacterial strains found in the middle ear cavity
are also generally found in the nasopharynx. This was
demonstrated:
By bacterial cultures of both middle ear effusion
(MEE) and nasopharyngeal secretions. Recently in
a multicentric study, 63.3% of 359 children suffering
from AOM had the same bacteria in the MEE and
in the nasopharyngeal secretions, either alone or
among other pathogens [l].The lack of sensitivity
(or exhaustivity) of nasopharyngeal sampling may
explain why the correlation between middle ear and
nasopharyngeal pathogens was not 100%~
By the analysis of outer membrane proteins and
lipo-oligosaccharides of non-typeable Haernophilur
infuenzue strains isolated simultaneously from the
nasopharynx and middle ear which showed that
these strains were generally identical in a given
individual [2].
With genetic fingerprinting [3].
Colonization of the nasopharynx
Faden et al. demonstrated that during episodes of AOM,
both the carriage rate and the absolute quantity of the

3s5
3S6 Clinical M i c r o b i o l o g y a n d I n f e c t i o n , Volume 3, S u p p k e m e n t 3
bacterial pathogens causing AOM, such as Haemophilus
infuenzae, Stveptococcus pneurnoniae and Moraxella ratav-
rhalis, increase in nasopharyngeal secretions [4].
The mechanisms of the colonization of the
nasopharynx are only partly understood. Bacterial
adherence to epithelial and pharyngeal cells depends on
several factors. The predilection of some strains of
bacteria such as S. pneurnoniae types 6A, 6B, 14, 19F
and 23F to induce AOM may be based on their
enhanced ability to adhere to and colonize the
nasopharynx, ascend the eustachian tube (ET), and
invade the middle ear [S].
The colonization of the nasopharynx depends also
on host and environmental factors. Differences in the
ability of bacterial pathogens to colonize the naso-
pharynx of different individuals may partially explain
why certain individuals become infected and others do
not. The colonization is clearly an age-dependent
phenomenon [6]. It is more important in infants
than in older children or adults. Shimamura et al. have
demonstrated that strains of H . infnenzae and S. pneu-
rnoniae adhered better to nasopharyngeal epithelial cells
of children than to those of adults and to those of
children with otitis media than in normal subjects [7].
Aniansson et al. have reported that the casein fraction
of the human milk inhibits the attachment of S.
pneumoniae and H . infnenzae to human respiratory
epithelial cells, suggesting a possible mechanism by
which breast-feeding may protect against AOM [8].
The attachment of H . infuenzae and S. pneumoniae,
which is a prerequisite to middle ear infection, was
significantly decreased in children demonstrating specific
secretory IgA in nasopharyngeal secretions compared
with children without such activity [7]. Lindberg et al.
demonstrated that the level of nasopharyngeal cyto-
kines, which may play a role in the regulation of the
local inflammatory response to bacteria, was lower in
children with recurrent AOM (rAOM) than in controls
[6]. From these data the authors hypothesize that rAOM
children have a local defective immune reactivity.
Colonization of the middle ear
Three mechanisms may explain how bacteria reach the
middle ear via the ET.
Aspiration
Negative middle ear pressure induced by sniffing opens
the ET and may induce an aspiration of pathogenic
bacteria into the middle ear [9].
lnje ction
Another possible way for bacteria to reach the middle
ear is by injection from the nasopharynx when its
pressure increases, as during nose blowing or sneezing.
Stenstrom et al. investigated the possible involve-
ment of a dysfunction of the ET in the etiology of
rAOM [9]. They tested the E T functions of children in
a healthy state and compared results of otitis-prone
children to those of children with no history of AOM.
They found that children with a history of AOM had
a lower residual middle ear pressure after closure and a
poorer active tubal function than healthy children. Yet
there is no clear evidence as to whether the dysfunction
is a primary cause or only secondary to otitis media.
Contiguous spread
The third hypothesis is that bacteria may spread
gradually from the nasopharynx to the middle ear. This
process is probably facilitated by viruses and/or by the
production of a number of exoproducts by respiratory
bacteria. These enzymes m o d i e receptors on the E T
cells, facilitating adherence and colonization as they do
on the pharyngeal cells.
Exposure of part or all of a S. pneumoniae adherence
receptor structure by the pneumococcal enzymes,
such as neuraminidase and glycosidase, may facilitate
colonization, invasion of the middle ear, and induction
of AOM. Linder et al. studied the evolution of the cell
surface carbohydrates in the E T and the middle ear
mucosa after intranasal inoculation of S. pneumoniae in
the chinchilla [S]. They demonstrated that such an
inoculation leads to the removal of the terminal sialic
acid, and also to the exposure of N-acetylglucosamine,
a component of a pneumococcal trisaccharide receptor
previously identified on human pharyngeal cells, which
progressed on the tubal surface from the nasopharyn-
geal to the tympanic orifice.
Bacteria present in the middle ear during AOM
In this paper only two points will be discussed: the
presence or absence of bacteria in middle ear effusion
(MEE) during current AOM and their load.
Sterile cultures
Bacteria are cultured from MEE obtained from children
with AOM in only 5 5 4 2 % of the cases [lo-131. In
the remaining cases, other microorganisms, such as
anaerobic bacteria or viruses, were regarded as the
causative agent(s) [10,14]. However, recently more
sensitive methods have revealed that culture-negative
samples of MEE may contain bacteria remnants. For
example, Virolainen et al., who have compared a
pneumococcal polymerase chain reaction (PCR) to
bacterial culture with 180 MEE samples of children
with AOM, have found a positive pneumolysin
PCK for 51 (28%) of 180 MEE samples, whereas
S. pneumoniae was cultured in only 33 (18%) samples
~51.
FranGois: N e w v i e w s o n t h e p a t h o g e n e s i s of acute o t i t i s m e d i a a n d i t s c o m p l i c a t i o n s
Bacterial load
The high bacterial load found in MEE is not specific
of an acute infection. Raisanen and Stenfors demon-
strated that the bacterial load of the MEE was
iionsignificantly different in AOM and otitis media
with effusion (OME), since the effusion was of the
niucopurulent type [13]. They have collected 49
samples from AOM and 128 samples from OME.
When MEE obtained during AOM was not sterile,
bacterial counts were in the range 106-10X/niL. When
niucopurulent MEE obtained during OME was not
sterile, the bacterial count was also between 106and 108
bacteria per mL, whereas in serous and mucoid effusions
the bacterial count was nil and 104-10s, respectively.
MIDDLE EAR REACTION TO ACUTE INFECTION
Structural changes
Otitis media is characterized by inflammation of the
middle ear. There is a continuum of the pathologic
changes, progressing from an acute to a subacute and
then to a chronic phase, in which irreversible tissue
damage is observed.
In the earliest stage of AOM, the middle ear
niucosa shows alterations characteristic of an inflam-
matory response, particularly dilatation and increased
permeability of capillaries, edenia of the laniina
propria, and leukocytic infiltration [I 61.
The histopathology of experimental purulent otitis
media caused by viable bacteria has been extensively
studied in rodents. In the chinchilla, 3 days after
inoculation of the middle ear space with viable S.
pneimoniae there was an acute inflammation of poly-
niorphonuclear leukocytes (PMNLs) in the subepithelial
space of the middle ear mucosa and subepithelial
edema [ 171. Epithelial nietaplasia, increasing edema,
and middle ear effusion containing PMNLs, were
apparent by day 7. In rats, 1 day after infection of the
middle ear with Staphylococcus sums there was an acute
inflammation of PMNLs and lymphocytes, and a
purulent discharge. Histopathologic studies revealed an
increase of the ciliated and secretory cells. Another
experimental study performed by Widemar et al.
in the rat showed that niast cells, usually located
beneath the outer epithelial layer, could be observed
thoughout the whole depth of the connective tissue
layer in cases ofpurulent otitis tnedia [18]. All mast cells
contained rather few granules, as if they had been
released shortly before the death of the animals. The
niast cell granules contain several mediators which may
initiate and sustain the inflammatory reaction
Effusion
Usually the MEE samples obtained during AOM
3S7
contain many PMNLs as mucopurulent MEEs of
OME, whereas inucoid MEE samples obtained during
OME contain very few cellular elements.
Apart froni these cells and bacteria, MEE contains
secretory products from glands of the middle ear
mucosa, and products from inflammatory cells or
bacterial organisms in the middle ear mucosa or in the
middle ear cavity [ 191.
Neutrophils release free radicals such as H202,
which may inhibit the cell nietabolism and thus
prolong the niiddle ear inflammatory response and
lead to chronic tissue damage. The effect of neutrophil
metabolic products on middle ear epithelial cells of the
chinchilla was studied in vitro by Kawana et al. [20].
They demonstrated that cell growth was inhibited by
activated neutrophils whereas tinstimulated neutrophils
did not affect the viability of these cells. Other in vitro
experiments with huniaii neutrophils showed that the
stimulation of neutrophils by bacterial components
modifies ion transport. This effect was inhibited by
catalase, which allowed Herinan et al. to advocate the
role of free radicals in the effect of neutrophils on
epithelial cells [2 11. Neutrophils also produce various
enzymes such as phospholipase and myeloperoxidase.
In the middle ear cavity, where inflammatory
effusions may stagnate, immune coniplexes may have
the potential to activate complement, release pharma-
cologically active mediators, and thus evoke persistent
inflammatory responses.
Inflammatory mediators
Infection and E T dysfunction induce secretion of
inflammatory mediators, which in turn cause increased
vascular permeability and stimulate epithelial secretory
activity, resulting in niiddle ear inflanirnation and
effusion.
Eicosanoids
Phospholipases release arachidonic acid (AA) froni the
membrane phospholipids. AA metabolites (such as
prostaglandins and leukotrienes) are continuously syn-
thesized by lipoxygenase and cyclo-oxygenase and are
continuously and rapidly destroyed [ 191. This activity
seems to be important during infection, since levels
of AA metabolites were higher in purulent than
in non-purulent (serous or mucoid) OM. The pre-
dominant AA metabolite in chinchilla middle ear
mucosa was hydroxyeicosatetraenoic acid in the puru-
lent O M group. These experimental data suggest that
steroids, which inhibit both lipoxygenase and cyclo-
oxygenase pathways, may prove to be more effective
diminishing the inflammatory process of AOM than
non-steroidal anti-inflammatory drugs, which inhibit
only the cyclo-oxygenase pathway.
358 Clinical Microbiology and Infection, Volume 3, Supplement 3
Cytokines
Cytokines are glycoproteins produced by many cells,
including lymphocytes and macrophages. Some cyto-
kines act as mediators of inflammation and regulators
of the immune response so that they may have an effect
on the outcome of the disease. Juhn et al. have analyzed
70 effusion samples from children who had O M for
two cytokines (IL-1p and TNF-a) and total colla-
geiiase [22]. The highest concentrations of all three
inflammatory mediators were found in purulent O M ,
and concentrations were higher in younger than in
older patients.
Histamine
Chonmaitree et al. have measured the levels of histamine
in 677 MEE samples from children with AOM [23].
Histamine levels were higher in bacteria-positive than
in bacteria-negative MEE samples. They were higher
in samples from patients with viral infection than in
those from patients with n o viral infection. Bacteria and
viruses together had an additive effect on histamine
content in MEE. Histamine concentrations in the
initial MEE sample tended to be higher in patients with
persistent otitis than in those with good response to
treatment. Kesults suggest that viruses or bacteria, or
both, induce histamine production, which leads to
increased inflammation in the middle ear.
MEE samples obtained during AOM always show
great amounts of inflammatory mediators. Yet it is not
known whether and to what extent inflammation of
the middle ear is beneficial for the recovery of AOM.
ELIMINATION OF THE EFFUSION AND/OR THE
BACTERIA
Some bacteria probably continuously reach the middle
ear cavity via the ET. But infection occurs only when
there is an important increase of the inoculum (as
shown in experimental models of AOM) or when
both non-specific and specific defense mechanisms are
impaired [24].
Non-specific factors
A variety of non-specific factors, present in the middle
ear, may play roles in defense against infection.
Mucociliary clearance
The epithelium of the E T and middle ear is lined with
mucus-producing ciliated cells that are equipped to
trap and expel small particles such as bacteria [25,26].
It is expected that any impairment of the mucociliary
clearance may induce a pooling of secretions in the
middle ear, facilitating multiplication of bacteria which
may lead to AOM. Ciliated dysfunction in the middle
ear may also contribute to the ongoing inflammatory
process in AOM by preventing efficient clearance of
infected or irritating middle ear fluid.
Impairment of mucociliary function may be due
to loss of cilia or ciliated epithelium, lowered ciliary
beating activity, changes in the amount or the com-
position of the mucus, or periciliary fluid. Many studies
have shown that infections by respiratory viruses are
associated with transient abnormalities of cilia in the
upper respiratory tract epithelium [25,27]. In vitro and
in vivo experimental studies in the guinea pig showed
that injection of Staphylococcus aureux into the middle ear
cavity modifies the ciliary beating activity [26]. Various
endotoxins and inflammation mediators were also
proven to affect the ciliary activity [28].
Chemotaxis and phagocytosis
Bacteria such as H. influeuzae, S. pneumoniae and
Staphylococcus aweus may be destroyed by phagocytosis.
This mechanism is perhaps impaired in children with
AOM. Pneumococcal capsular polysaccharide was
detected in about half of the effusion samples from
children with acute pneumococcal O M [29]. These
subcellular components, if retained in the middle ear,
may sustain the inflammation.
In fact, chemotaxis and phagocytosis were studied
mainly in OME. Complement-activation products,
which comprised only a small fraction of the chemo-
tactin mediators, are present in MEE of children with
O M E [17]. The presence of live and dead Staphylococcus
aureus bacteria was almost the same in peripheal blood
neutrophils and in middle ear neutrophils and macro-
phages of children with O M E [30]. But it appeared
that S. pneurnoniae was not phagocytosed well by either
peripheral blood or middle ear phagocytes. Bernstein
hypothesized that this may be related to the absence of
specific antibody in the patient’s serum at the time of
infection.
Miscellanous
The network of fibrin present in middle ear effusions,
particularly in purulent effusions, restricts movement of
organisms and facilitates phagocytosis 1251. Destruction
of white blood cells and cells lining the mucosa
produces lactic acid and a decrease in p H sufficient to
kill or inhibit growth of many bacteria [25]. MEE also
contains proteins which inhibit bacterial proliferation,
either by iron chelation (transferrin) or by their
own bactericidal activity. Lysozyme, the major serum
murein hydrolase, is present in MEE. It appears to be
derived from PMNLs as well as frotn other sources,
perhaps lysozyme-secreting cells in the middle ear and
E T epithelium [25].
FranGois: N e w v i e w s o n t h e p a t h o g e n e s i s o f a c u t e o t i t i s m e d i a a n d i t s c o m p l i c a t i o n s
Specific immune response
The role of systemic and local immune responses in
the pathogenesis and natural history of, and the
mechanism of recovery from, AOM is not completely
determined.
Normal middle ear niucosa contains only a few
ininiunoconipetent cells 1121. During infection, how-
ever, the middle ear inucosa appears to be transformed
into a secretory organ capable of immunologic
reactivity. There is a local synthesis of serum type
antibodies and secretory component in the plasma
cells and epithelial cells of the middle ear mucosa,
respectively.
Experimental studies
Several investigations in experimental animal models
have been carried out to examine the role of local and
systemic immunity in the clearance of bacteria from the
mucosa [h].
Two longitudinal studies of the efficacy of anti-
biotics in the treatment of pneumococcal AOM in a
chinchilla model suggest that early administration of
systemic antibiotics interferes with the local immune
niechanisni in the middle ear 1311. The investigators
suggested that systemic humoral immunity probably
played little part in protecting the middle ears of these
animals when directly inoculated with S. pneumoniae.
It seems that bacterial antigens must reach imniuno-
competent cells at the local immune site for some local
protective immune response to occur.
A similar experiment with non-typeable H. irzfu-
enzae demonstrated that the protection against a second
AOM with the same strain was bilateral, which suggests
that it depends on a systemic immune response [32].
Clinical studies
A variety of immunologic disturbances have been pro-
posed as contributing to the development of rAOM.
It has been suggested that proneness to otitis may
be due to an impaired response to protein antigens from
non-typeable N.injuenzae [12]. In contrast, the role of
the local immune response is predominant in recurrent
pneumococcal O M , as it may occur even in the
presence of serum type-specific antibodies [33].
Karjalainen et al. demonstrated the appearance
of MEE antibodies against the three most common
bacteria causing AOM before the thirteenth hour in
children suffering from AOM 1121. About one third of
the children also had specific antibodies in MEE to
bacteria other than the causative one. Even culture-
negative effusions contained specific antibodies. These
findings suggest that any infection in the middle ear
may evoke local antibody production of lymphocytes
sensitized to antigens of the most frequent bacterial
3s9
pathogens in AOM. The MEE levels of specific anti-
bodies seemed to correlate well with the corresponding
serum antibody levels. Thus, it seems that the MEE
antibodies in the early phase of AOM derive mainly
from the serum by transudation 1121. The antibodies in
the MEEs against H . injuenzae and 111.catarrlialis usually
consisted of both IgG- and IgA-class antibodies, as well
as secretory antibodies, whereas antibodies against
S. pneurnoniae consisted mainly of IgG and IgM types.
Secretory type IgA-class antibodies against S. pneu-
rnoniae and serum type IgA-class antibodies against
S. pneunzoniae were more often found in the later
phase ofAOM. It therefore seems that the IgA response
to S. pneunzoniae develops slowly during AOM and
is at least partly mediated by local immune mechanisms.
These results are in contradiction to other data
which demonstrated that in the case of established
AOM, elimination of the effusion was faster at
higher concentrations of pneumococcal antibodies
[34]. Furthermore, Prellner et al. noted that rAOM
of children whose cord serum contained a low level of
specific IgG antibodies against pneutnococcal types 6A
and 19F was mainly due to S. pnetrinoniae (although
their cord serum concentrations of total IgG or the
subclasses IgGl or IgG2 were normal) [34].
PHVSIOPATHOLOGV OF THE COMPLICATIONS
OF AOM
Chronic otitis media or otitis media with effusion
Clinically, 5-10% of episodes of AOM lead to chronic
O M E 1171. Animal studies have shown that chronic
O M E lies on a continuum with acute suppurative O M
1161.
There is accumulative evidence that OME is not
only due to tuba1 dysfunction and that bacteria or
subcellular bacterial components play a role in the
transition from AOM to chronic O M E (291. Recent
studies in chinchillas suggest that bacteria, as well as cell
envelope and cell wall components released during
bacterial lysis, may contribute to OME [17]. Less than
half of the effusions from children with chronic O M E
yield bacteria on culture, but endotoxin was detected
in 80% of effusions, and pneumococcal capsular poly-
saccharides were detected in 33% of 202 sterile
middle ear effusions from children with chronic OME.
Purified pneumococcal cell wall alone caused patholog
similar to that seen in ears inoculated with killed,
encapsulated pneuniococci. Unencapsulated pneumo-
cocci caused more inflammation than encapsulated
ones. Bacterial lipopolysaccharides (i.e. endotoxin) of
Gram-negative organisms may also be responsible for
persistent inflammation of the middle ear. Researchers
have demonstrated that as little as 1 ng of endotoxin
3510 C l i n i c a l M i c r o b i o l o g y a n d I n f e c t i o n , V o l u m e 3, S u p p l e m e n t 3
from H. inzuenzae induces severe inflammatory changes
in the niucoperiosteum of experimentally inoculated
chinchillas [17]. In the clinical setting, endotoxin has
been detected in MEE in substantial numbers ofpatients.
Mastoiditis
The proximity of the mastoid to the middle ear
cleft suggests that most cases of suppurative O M are
associated with inflammation of the mastoid air cells.
Clinically significant mastoiditis is rare since the intro-
duction of antimicrobial agents. Infection in the
mastoid proceeds after middle ear infection through the
following stages: (1) inflammation of the mucosal lining
of the mastoid cells; (2) accumulation ofpus in the cells;
(3) impairment of the bony walls due to both pressure
of purulent exudate and ischemia, and infection of the
periosteum from the mastoid air cells by venous channel;
(4) destruction of the cell walls and coalescence of
adjacent cells; and (5) escape of pus into contiguous
areas. This process may halt at any stage with sub-
sequent resolution [25]. Since the introduction of
systematic treatment of AOM by antibiotics, the disease
rarely oversteps stage 2.
S. pneumoniae is by far the most frequently isolated
organism in acute mastoiditis, followed by group A
streptococci [35]. It remains unexplained why such
organisms, susceptible to most antibiotics used for
treatment of any infection of the upper respiratory tract
in children, are consistently isolated in acute mastoiditis
while the more resistant species H . inzuenzae is rarely
found. Some authors hypothesized that H . influenzae
causes disease of mucous and serous membranes but is
less likely to attack bone.
Acute mastoiditis can occur without any recent
symptom of AOM. It still remains unknown if the
AOM in these cases takes a subclinical course or if
failure in diagnosis plays a role, especially in young
infants [35].
Sensorineural hearing loss
Although rare, sensorineural hearing loss is a possible
complication of AOM [36]. The involvement of the
inner ear during AOM seems to be through the round
window membrane. It has been demonstrated that the
thickness of the round window membrane is increased
in temporal bones from subjects with otitis media as
compared to non-otitis media temporal bones [37].
This increased thickness has been suggested to serve as
a protective mechanism for the inner ear, inhibiting the
passage of toxins. Evidence supporting this hypothesis
includes a decreased permeability of the round window
membrane in temporal bones with experimentally
induced chronic OM, and those at the resolved stages
of purulent OM [38,39].
Meningitis
Haernophilus in fluenzae
A number of experiments in the rat model and
observations in humans led to the conclusion that
H . inzuenzae meningitis occurs by the hematogenous
route, rather than by contiguous spread from the naso-
pharynx or the middle ear. Bacteremia during acute
infection with H . injuenzae type b may lead to a
central nervous system invasion when the number of
organisms is high (>lo3 colony forming units/mL)
1401. The intranasal route has been shown to lead to
H . injuenzae meningitis by hematogenous dissemina-
tion in the infant rat model and the infant primate
model [41].
Streptococcus pneumoniae
In contrast, pneumococcal meningitis secondary to
AOM is secondary to bacterial passage from the middle
ear to the inner ear. This has been demonstrated in
rodents (24,381. All gerbils whose ear was inoculated
with lo4 CFU of S. pneumonia presented AOM, and
76% developed a brain tissue infection in 25 cases,
whereas only 42% had a bacteremia. The degree of
dissemination was markedly related to bacterial counts
in MEE. This result suggests a direct extension from
the middle ear to the brain. Anatomic bacterial path-
ways promoting meningeal extension of pneumococcal
AOM remain to be determined. An experimental study
in the chinchilla model of AOM supports the hypo-
thesis that infection spreads along neuronal preformed
pathways after inner ear penetration via the round
window membrane [38].
CONCLUSION
AOM is due to the bacterial colonization of the
middle ear by nasopharyngeal bacteria. Recent studies
have shown that this colonization is facilitated by the
enhancement of bacterial adherence on the E T lining
by viruses and/or bacterial enzymes. The mucociliary
clearance, which seems to be the main protective
mechanism of the middle ear against infection, appears
to be impaired during AOM by combined respiratory
viruses, endotoxins and inflammatory mediators. The
infection prompts an inflammatory response with
increased permeability of capillaries and leukocytic
infiltration which participate in the fight against
infection by both phagocytosis and immune responses.
Clinical and experimental data have demonstrated that
an incomplete clearance of bacteria may be responsible
for persistence of inflammation in the middle ear
which leads to chronic OME, the main complication
of AOM.
Franqois: New views o n t h e pathogenesis of acute otitis media and its cornulications
References
1 . Lenoir G, Gehanno €? Apport du preltvenient nasopharyngk
dam la documentation bacteriologique des otites moyeniies
aigues de l’enfant. Lettre de I’lnfectiologue 1994; 9(suppl
18): S23-5.
2. Murphy TF, Bernstein J M , l1ryja IIM, Campagnari AA,
Apicella MA. Outer membrane protein and lipooligo-
saccharide analysis of paired nasopharyngeal arid middle ear
isolates in otitis media due to noii typable Huenrophilrrr
irg‘luenzae: pathogenetic and epidemiological observations.
J Infect Dis 1987; 156: 723-31.
3 . Loo? UG, Uernstein JM, Dryja I l M , Murphy TF, Dickinsoii
111’. Determination of the epideniiology and traiisniission
of nontypable Hurmnpliilrrs inflrrenzue in children with otitis
niedia by comparison of total genoniic D N A restriction
fingerprints. Infect Immun 1989; 57: 2751-7.
4. Faden H , Stanievich J. Brodsky L, Bernstein J, Ogra PL.
Changes in nasopharyngeal flora during otitis media of
childhood. Pediatr Infect Dis J 1990; 9: 623-6.
5. Linder TE, Daniels R L , Lim DJ, Denlaria TE Effect of
intranasal inoculation of Streptocorcur pnrrrrnowiile on the
structure of the surface carbohydrates of the chinchilla
eustachiaii tube and middle ear mucosa. Microb Pathogen
1994; 16: 435-41.
6. Lindberg K, Rynneldagoo B, Sundqvist KG. Cytokincs in
nasopharyngeal secretions, evidence for defective IL-1-beta
production in children with recurrent episode5 of acute otitis
media. C h i Exp Irnniunol 1994; 97: 396-402.
7. Shimamura K, Shigemi H, Kurono Y, Mogi G. The role
of bacterial adherence in otitis media with effurion. Arch
Otolaryngol Head Neck Surg 1990; 116: 1143-6.
8. Aniansson G, Andersson B, Lindstedt R, Svanborg C. Anti-
adhesive activity of human casein against Sfrepfococcirs pwcn-
nioninr, aiid Huermiphilirs i@rienzar. Microb Pathogen 1990;
8: 315-23.
9. Steiistroni C, Bylander-Groth A, Ingvarsson L. Eustachian
tube fuiiction in otitis-prone and healthy children. hit J
Pediatr Otoi-hinolaryngol 1991; 21: 127-38.
10. Choniiiaitree T, Howie VM, Truant AL. I’reseiice of
respiratory viruses in middle ear fluids and nasal wash
specimens from children with acute otitis media. Pediatrics
1986; 77: 698-702.
11. Trujillo H, Callejas R, Mejia GI, Castrillon L. Bacteriology
of nllddle ear flnid specimens obtained by tympanocentesis
from 1 1 1 Colombian children with acute otitis media.
Pediatr Infect I l i r J 1989; 8: 361-3.
12. Karjalaineii H, Koskela M , Luotonen J, Herva E, Sipila
P. Antibodies against Streptococcns pneurnorzia, Haer~rtipliilris
irlfiiwnzue m d Brunhnmelln catarrhillis in nnddle ear effusion
during early phase of acute 0 t h media. Acta Otolaryngol
(Stockh) 1990; 109: 111-18.
13. Kaisanen S,Stenfors LE. Bacterial quantification, a necessary
complement for the comprehension of middle eJr inflarii-
inations. hit J I’ediatr Otorhinolaryrigol 1992: 23: 117-1 24.
14. Brook I. Pathogenicity and therapy of anaerobic bacteria in
upper respiratory tract infections. Pediatr Infect I3is J 1987;
6: 131-6.
15. Virolainen A, Salo P, Jero J, Kariiia l’, Eskola J. Leinonen M .
Comparison of PCR assay with bacterial cultiire for detect-
3S11
ing Strepfocorcrrs pnerrrm~niue in middle ear fluid of children
with acute otitis media. J Clin Microbiol 1994; 32: 2667-
70.
16. Wright CG, Meyerhoff WL. Pathology of otiti5 media. Ann
O t o Rhinol Laryngol 1994; 103(suppl 163): S24-6.
17. Ripley-Petzoldt ML, Giebirik GS, Juhn SK, Aeppli D,
Toinasz A, Tuomanen E. The contribution of pneumococcal
cell wall to the pathogeneTi5 of experinientd otitis media.
J Infect I1is 2988; 157: 245-55
3 8. Widemar L, Hellstroni S, Stenfors LE. Different structural
changes in membrana Shrapnellr in sero~isand purulent otitis
media, an experimental study in the rat. Acta Otolaryngol
(Stockh) 1986; 102: 266-73.
19. Jung TTK. Prostaglandins, leukotrienes, and other aracht-
donic acid metabolites in the pathogenesis of otitis rnedi‘i.
Laryngoscope 1988; 98: 980-93.
20. Kawana M , Kawana C, Amesara R, Juhn SK, Giebink GS.
Neutrophil, oxygen metabolite inhibition of cultured
chinchilla middle ear epithelial cell growth. Ann Otol
Rhinol Laryngol 1994; 103: 812-16.
21, Herman P, Tran Ba H ~ i y€? Mod& experimental d’inflan-
mation de l’oreille inoyenne. In Magrian J, ed. L‘otite
chroiiique. Paris: Arnette, 1995: 27-30.
22. Juhn SK, Garvis WJ, Le CT, Lees CJ, Kim CS. Ileterniining
otitis mcdia severity froni middle ear fluid analysis. Ann Otol
Rhino1 Laryngol 1994; 103(suppl 163): S43-5.
23. Chonmaitree T, Pate1 JA, Lettbrown MA, Uchida T,
Garofalo R , Owen MJ, Howie VM. Virus and bacteria
enhance histamine prodiiction i n iniddle ear fluids of children
with acute otitis media. J Infect Dis 1994: 169: 1265-70.
24. Muffat-Joly M , Barry 13, H h i n 11, Fay M, Gehaiino 1’.
Otogenic tneningoeiicephalitis induced by Streptococctts
prier4rm~riiaein gerbils. Arch Otolaryrigol Head Neck Surg
1994; 120: 925-30.
25. Bluestone U l , Klein JO. Mastoiditis. In Dlucstone C11,
Stool SE, eds. Pediatric otolaryngology. Philadelphia: W B
Saunders, 1990: 546-52.
26. Ohashi Y,Nakai Y,Kihara S, Ikroka H . Effkcts of Staphylo-
cocci~~uiit’eirs on the ciliary activity of the middle ear lining.
Ann Otol Rhinol Laryngol 1987; 96: 225-8.
27. Chung M H , Grifith SR, Park KH, Lim DJ, L k Maria TI?
Cytological and histological change5 in the middle ear after
inoculation of influenza A virus. Acta Otolaryngol (Stockh)
1993; 113: 81-7.
28. Ohashi Y,Nakai Y, Ikeoka H, Kochinii H , Eyaki Y. Effects
of bacterial endotoxin on the ciliary activity in the in vitro
middle ear tnucosa. Acta Otol.iryngo1 (Stockh) 1987; 104:
495-9.
29. Karma P, Sipila P, Virtanen T, Luotonen J , Sipila M , Haiva
VM. T’neuntococcal bacterioloby after pneumococcal otitis
media with special reference to pneumococcal antigens. Int
J l’ediatr Otorhitiolaryiigol 1985; 1(I: 181-90.
30. Bernstein JM. Reccnt advance\ in immunological reactivity
in O M E . In: Proceedings of- Inpharzain medical forum 0 1 1
otitis media with effusion (Lugano April 1983) 1984: 5:
25-31.
31. Marshak G, Chitekin El, l h y l e WJ, Shiffnian C;, Kohn 1 ) I l
Recurrent pneuinococcal otitis iiiedia i n the chinchilla, a
longitudinal study. Arch Otolai-yngol 1981; 107: 532-9.
3S12 Clinical Microbiology and Infection, Volume 3, Supplement 3
32. Karasic RB,Trumpp CE, Gnehm HE, Rice PA, Pelton
SI. Modfication of otitis media in chinchillas rechallenged
with nontypable Haemophilus infuenzae and serological
response to outer membrane antigens. J Infect Dis 1985; 151:
273-9.
33. Gray RM, Dillon H C Jr. Epidemiological studes of Strepto-
coccus pneumoniae in infants: antibody to types 3,6,14 and 23
in the first two years of life. J Infect Dis 1988; 158: 948-55.
34. Prellner K, Kalm 0, Harsten G, Heldrup J, Oxelius VA.
Pneumococcal serum antibody concentrations during the
first three years oflife: a study of otitis-prone and non-otitis-
prone children. Int J Pediatr Otorhinolaryngol 1989; 17:
267-79.
35. Hoppe JE, Koster S, Bootz F, Niethanlmer D. Acute
mastoidtis, relevant once again. Infection 1994; 22: 178-
82.
36. Morizono T, Giebink GS, Paparella MM, Sikora MA, Shea
DS. Sensorineural hearing loss in experimental purulent
otitis media due to Streptococcuspneumoniae. Arch Otolaryngol
Head Neck Surg 1985; 111: 794-8.
37. Sahni RS, Paparella MM, Schachern PA, Goycoolea MV, Le
CT. Thickness of the human round window membrane in
different forms of otitis meda. Arch Otolaryngol Head
Neck Surg 1987; 113: 630-4.
38. Schachern PA, Paparella MM, Hybertson R, Sano S, Duvall
AJ. Bacterial tympanogenic labyrinthitis, meningitis, and
sensorineural damage. Arch Otolaryngol Head Neck Surg
1992; 118: 53-7.
39. Ikeda K, Sakagami M, Morizono T, Juhn SK. Permeability
of the round window membrane to middle-sized molecules
in purulent otitis media. Arch Otolaryngol Head Neck Surg
1990; 116: 57-60.
40. La Scolea LJ, Rosales SV, Welliver R C , Ogra PL.
Mechanisms underlying the development of meningitis or
epiglottitis in children after Haemophilus infuenzae type b
bacteremia. J Infect Dis 1985; 151: 1162-5.
41. Moxon ER, Glode MP, Sutton A, Robbins JB. The infant
rat as a model ofbacterial meningitis. J Infect Dis 1977; 136:
S186-90.