chronic otitis

7/28/2019 Chronic Otitis

1/16

section four

Middle Ear, Mastoid, and Temporal Bone

cHAPter one HunDreD AnD tHirtY-nine

ch o M d a, Ma da d PRichard A. Chole

Holger H. Sudhoff

Key Points

Otitis media is one of the most common diseases of childhood, and may result in complications including acute and chronicmastoiditis, petrositis, skull base osteomyelitis, intracranial infection, and the sequelae of early childhood auditory deprivation.

Acute otitis media may result in persistent otitis media with effusion, which is now recognized as the leading cause of childhoodhearing loss.

Aural cholesteatomas are epidermal inclusion cysts of the middle ear or mastoid, and are classi ed as congenital and acquired. Cholesteatomas can be eradicated from the temporal bone only by surgical resection. Osteoclastic bone resorption in chronic otitis media with or without cholesteatoma is stimulated by various factors, including

in ammation, local pressure, keratin, speci c cytokines, and bacterial toxins. The mechanism of bone resorption in chronic otitis media with and without cholesteatoma is an increased number and activity of

osteoclasts. Tympanosclerosis is tympanic membrane or middle ear hyalinization. It results from chronic in ammation or trauma, but it often

leads to conductive hearing loss caused by ossicular xation. Petrous apicitis is an extension of infection from the mastoid air cell tract into a pneumatized anterior or posterior petrous apex. The management of petrous apicitis is directed toward control of the infection, such as topical and systemic antibiotic management

and surgical approach through many surgical routes.

Otitis media is the most common disease of childhood after viral upperrespiratory infections. The acute bacterial infection occurs in 80% of children 1 to 6 years old, and it is the disease most frequently managed

with antibiotics in the United States. The infectious and noninfectiouscomplications of otitis media in childhood may result in serious mor-bidity. Later in life, the infectious complications, including acute andchronic mastoiditis, petrositis, and intracranial infection, still occurdespite the widespread use of antibiotics for this disease. The noninfec-tious sequelae, including chronic perforation of the tympanic mem-brane, ossicular erosion, labyrinthine erosion, and tympanosclerosis, aremajor causes of hearing loss throughout the world.

Some cases of acute otitis media (AOM) result in persistent otitismedia with effusion (OME), which is recognized as the leading causeof childhood hearing loss. The exact cause of OME is unclear; however,more recent data suggest that re ux of gastric contents may be associ-ated with OME in children.1-4 Although eustachian tube dysfunctionalone may lead to effusion of the middle ear, there is mounting evidencethat most cases of OME occur as a sequelae of AOM, or at least sharethe same etiologic factors. Speci c causes can be identi ed in many cases of adult-onset otitis media, such as paranasal sinus disease,nasopharyngeal carcinomas and tumors, and postradiation sequelae.

In most children, AOM and OME subside spontaneously or aftermedical intervention. It is unknown how many children with OMEeventually have complications. The sequelae of otitis media can be

considered in two broad categories: (1) direct destructive effects of thelocalized process, and (2) the effects of auditory deprivation duringearly childhood. Otitis media may be complicated by acute or chronicperforation of the tympanic membrane, acute mastoiditis, middle earatelectasis, adhesive otitis media, tympanosclerosis, ossicular erosion o

xation, petrous apicitis, cholesteatoma, chronic otomastoiditis, laby-rinthitis, facial paralysis, and intracranial infection. There is evidencethat sensorineural hearing loss may result from chronic otitis media

with or without cholesteatoma. There is also evidence that the auditory deprivation associated with childhood otitis media may lead to indirectsequelae, such as language and speech delays.

Effects on Mastoid PneumatizationIt has been observed that patients with a history of chronic OME havemore sclerotic mastoids with decreased pneumatization compared withhealthy subjects. There are two theories to explain this observation: (1)the hereditary theory, which states that children with hypoaeration of the mastoid are prone to OME, and (2) the environmental theory,

which states that chronic OME results in hypopneumatization of themastoid.5 Although measurable correlations between mastoid hypocel-lularity and OME,6,7 and between the length of the mastoid process orthe degree of pneumatization and an abnormal eardrum8 have beenproven, a cause-and-effect relationship is unclear. Available evidencegenerally supports the concept that chronic in ammation in early
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


2/16

1964 Part 7 n o l gy, n - l gy, a d sk ll Ba s g y

treated with tubes developed severe atelectasis.12 It may be that repeatedbouts of AOM lead to weakening and thinning of the membrane,

which allows atelectasis. Sad and Berco13 showed destruction of thecollagen-containing brous layer of the tympanic membrane in someears with recurrent infection. Collagen destruction within the tympanicmembrane may lead to another complication of OMEtympanoscle-rosis. Sad and Berco13 and Tos and Poulsen14 described four stages of tympanic membrane retraction: stage I, retracted tympanic membrane;stage II, retraction with contact onto the incus; stage III, middle earatelectasis; and stage IV, adhesive otitis media (Fig. 139-3).

Middle ear atelectasis may be reversible with ventilating tubes

Sad15

showed that ventilating tubes improved the state of atelectaticears. Graham and Knight16 reported three cases in which atelectatictympanic membranes were restored to their normal position by admin-istration of nitrous oxide during anesthesia and insertion of a ventilat-ing tube.

Atelectasis and adhesive otitis media usually coexist with OME,although OME may resolve in these ears, allowing aeration of theattic and mastoid, but leaving a collapsed middle ear. In extreme cases,

when hearing loss or ossicular erosion occurs, a myringoplasty forthe reinforcement of atelectatic tympanic membrane may be indi-cated.6,17 Cholesteatomas may originate from deep retraction pocketsin which desquamated keratin debris would not be cleared into theear canal.18,19 These retraction pockets may occur in the pars tensa or pars accida of atelectatic ears, and should be considered precur-sors to cholesteatomas (see discussion of cholesteatoma). Nonpneu-matized mastoids may have a limited ability to buffer pressurechanges and can manifest as an atelectasis, a retraction pocket, or acholesteatoma.20

Chronic Otitis Media with Cholesteatoma Aural cholesteatomas are epidermal inclusion cysts of the middle ear ormastoid (in the case of a retraction pocket cholesteatoma, the cystopens into the external auditory canal). They contain the desquamateddebris (principally keratin) from their keratinizing, squamous epitheliallining. Cruveilhier21 rst described aural cholesteatoma as a pearlytumor of the temporal bone. The term cholesteatoma,coined by theGerman physiologist Mller22 in 1838, is a misnomer because thisentity does not contain cholesterol; the white-yellow keratin akesfound within cholesteatomas grossly resemble cholesterol crystalsCholesteatomas of the temporal bone may be congenital or acquired.

childhood may lead to new bone formation within the middle ear andmastoid and, subsequently, decreased size of mastoid air cells. Shatzand Sad9 measured the distance from the lateral sinus to the externalauditory canal and found it to be signi cantly smaller in patients withsclerotic mastoids; they believed that this nding supported the hered-itary theory because it was unlikely that otitis would change the posi-tion of the lateral sinus. A chronically in amed mastoid in a young child may not develop normally, however.

Hasebe and colleagues10 carried out a study to establish which typeof cholesteatoma is controllable by conservative treatment from theviewpoint of mastoid ventilation. Their clinical observations suggest

that progressiveness of cholesteatoma is related to the ventilatory condi-tions in the mastoid, rather than eustachian tube function, and thatconservative treatment may be effective if ears with cholesteatoma haveaeration in the mastoid.

Middle Ear Atelectasis and AdhesiveOtitis Media

Middle ear atelectasis (Fig. 139-1) is thought to result mainly fromlong-standing eustachian tube dysfunction. One of the main functionsof the eustachian tube is ventilation of the middle ear and mastoid.Opening of the eustachian tube allows exchanging of gases and equal-ization between the environment and middle ear. The middle ear gasesalso are exchanged with the middle ear mucosa. Bilateral diffusionbetween the middle ear cavity and the blood may be an importantfactor in middle ear atelectasis because the gas composition of themiddle ear basically resembles that of venous blood.11

If the atelectasis develops, the tympanic membrane becomesretracted onto the promontory and the ossicles of the middle ear. Inatelectatic ears, the middle ear space is partially or completely obliter-ated, but the tympanic membrane is not adherent to the medial wallof the middle ear, and the mucosal lining of the middle ear is intact.In contrast, adhesive otitis media exists when the middle ear space istotally obliterated, and the tympanic membrane is adherent to theossicles and promontory; mucosal surfaces are not present. Retractionof the tympanic membrane may lead to erosion of the long process of the incus and the stapes suprastructure (Fig. 139-2).

Not all patients with chronic OME develop atelectasis; in mostpatients with OME, retraction of the tympanic membrane is limited.In patients with bilateral OME, 1.5% of untreated ears and 2% of ears

Figure 139-1. Middle ear atelectasis. Figure 139-2. Middle ear atelectasis with ossicular erosion.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


3/16

Chapter 139 n ch o M d a, Ma d , a d P 1965

until proved otherwise. Occasionally, a cholesteatoma cannot be seenotoscopically, but is discovered during tympanomastoid surgery.

The exact prevalence of cholesteatoma is unknown. In 1978, there were 4.2 hospital discharges per 100,000 population with choles-teatoma.25 In addition, there were 13.8 hospital discharges per 100,000population with chronic otitis media without cholesteatoma. Studiesby Harker23 documented an annual incidence of 6 cholesteatomas per100,000 population. Tos 26 found an annual incidence of 3 cholesteato-mas in children and 12.6 cholesteatomas in adults per 100,000 popu-lation. In the human temporal bones with chronic otitis media,cholesteatoma was observed in 36% of ears with perforations and in4% of ears without the perforated tympanic membranes.27

Pa h gCongenital cholesteatomas, by de nition, originate from areas of kerat-inizing epithelium within the middle ear cleft. Michaels28 showed that

Acquired cholesteatomas are the consequence of OME, AOM, or both. An understanding of the pathogenesis and pathophysiology of auralcholesteatoma is particularly important because the destructive natureof this entity is responsible for much of the morbidity associated withchronic otitis media. The propensity of cholesteatomas to erode boneand the lack of effective, nonsurgical management add importance tothe understanding of this disease.

D agThe diagnosis of aural cholesteatoma is made on otoscopic examina-tion, including endoscopic and microscopic evaluation or surgicalexploration. Special imaging procedures, such as high-resolution com-puted tomography (CT) and magnetic resonance imaging (MRI), may suggest the presence of cholesteatomas within the temporal bone, andmay be used to complement the clinical examination. High-resolutionCT is useful for operative planning and is recommended for all revision

mastoid operations. The symptoms of cholesteatoma vary; some cho-lesteatomas are asymptomatic, whereas others become infected andrapidly cause bone destruction. Some patients seen by the physicianshow slowly progressive conductive hearing loss, and frequently patients

with cholesteatomas consult a physician because of chronic otitis withpurulent otorrhea. The otorrhea from an infected cholesteatoma often is malodorous because of the frequent infection with anaerobicbacteria.23

Patients with infected cholesteatoma occasionally are misdiag-nosed as having external otitis. Careful follow-up evaluation and thor-ough canal dbridement of a patient with otorrhea are mandatory because the cholesteatoma may not be evident during an acute are-up.Some patients have signs and symptoms of the complications of a cholesteatoma: vertigo and hearing loss caused by a labyrinthine stula,facial nerve paralysis, or intracranial infection.

The otoscopic appearance of an aural cholesteatoma also varies. A typical attic retraction cholesteatoma (Fig. 139-4) appears as a defectof variable size adjacent to the posterosuperior portion of the tympanicmembrane. The center of the defect contains keratin debris (primary acquired cholesteatoma). In other patients, keratinizing epithelium hasmigrated through a perforation into the middle ear (secondary acquiredcholesteatoma) (Fig. 139-5). Cholesteatomas sometimes appear behindor within an intact tympanic membraneso-called congenital choles-teatoma (Fig. 139-6). These cholesteatomas have been considered tobe congenital, but more recent experimental evidence raises the possi-bility that they may originate during an in ammatory process.24 Aninfected cholesteatoma sometimes manifests as an aural polyp. Thesepolyps actually are granulation tissue at the junction between aneroding cholesteatoma and bone. The presence of an aural polyp in a chronically infected ear should be considered to be a cholesteatoma

Normal Stage IRetraction

Stage IISevere retraction

Stage IIIAtelectasis

Stage IVAdhesive otitis

Figure 139-3. Four stages of middle ear atelectasis. (Adapted from Sad J, Berco E. Atelectasis and secretory otitis media. Ann Otol Rhinol Laryngol .1976;85(Suppl 25):66.)

Figure 139-4. Primary acquired cholesteatoma in the region of thepars accida with scutum erosion.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


4/16


mastoid extension; and stage IV, mastoid involvement. They showed acorrelation between stage and risk of residual disease; stage IV carriea 67% risk of residual cholesteatoma.

The pathogenesis of acquired cholesteatoma has been debated formore than a century. There are four basic theories of the pathogenesisof acquired aural cholesteatoma: (1) invagination of the tympanicmembrane (retraction pocket cholesteatoma), (2) basal cell hyperplasia,(3) epithelial ingrowth through a perforation (the migration theory),and (4) squamous metaplasia of middle ear epithelium (Fig. 139-7).

Additionally, Sudhoff and Tos19 proposed a combination of the invagi-nation and basal cell theories as an explanation for retraction pocketcholesteatoma formation.

Invagination Theory The invagination theory 5 of the genesis of cholesteatoma generally isregarded as one of the primary mechanisms of the formation of atticcholesteatomas. Retraction pockets of the pars accida deepen becauseof negative middle ear pressure and possibly repeated in ammation (seeFig. 139-2). As the retraction pocket deepens, desquamated keratincannot be cleared from the recess, and a cholesteatoma results. Theorigin of such retraction pocket cholesteatomas is thought to be eus-tachian tube dysfunction (or OME) with resultant negative middle earpressure (ex vacuotheory). Usually, the pars accida, being less brousand less resistant to displacement, is the source of the cholesteatoma.

The result of this type of cholesteatoma (so-called primary acquired cholesteatoma) is an apparent defect in the posterosuperiorquadrant of the tympanic membrane and erosion of the adjacent canal

wall.31 Although these defects have the appearance of a marginal per-foration, it is not a perforation, but rather an invagination. Sad20 showed that epithelial migration patterns within attic retraction pocketsare altered. This failure of epithelial migration may allow the accumu-lation of keratin within a retraction pocket, with subsequent enlarge-ment merely from the accumulation of keratin within a relatively closedspace. This theory has been supported by the experimental creation of retraction pockets by use of the eustachian tube obstruction,32 andexternal auditory canal ligation.33

Ruah and colleagues31 suggested that in ammation of middle earand persistent mesenchyme lead to a greater in ammatory reaction inpars accida and posterosuperior quadrant of the tympanic membraneof the human temporal bones with serous and purulent otitis media in

children. These ndings support this theory of primary acquired choles-teatoma in children. Retraction pockets are considered as precursors tocholesteatoma. Bacteria can infect the keratin matrix, forming bio lmsleading to chronic persistent infection. The presence of bacterial bio-

lms in the cholesteatoma matrix may lead to epithelial proliferationand invasion of the cholesteatoma.34,35

Basal Cell Hyperplasia Theory Another possible mechanism for the histogenesis of cholesteatoma wassuggested by Lange in the 1920s.49 In this theory, he proposed thatepithelial cells (prickle cells) of the pars accida could invade the subepithelial tissue by means of proliferating columns of epithelial cellsNearly 40 years later, Ruedi50 supported this hypothesis with clinicaland experimental evidence. For epithelium to invade into the lamina propria, the basal lamina (basement membrane) should be altered. Basallamina disruptions now have been documented in human51,52 andanimal53 cholesteatomas.

Huang and colleagues54 and Masaki and coworkers55 providedexperimental support of this theory by showing that epithelial ingrowthfrom the tympanic membrane can be induced by instillation of propyl-ene glycol into the middle ear of chinchillas. These basal lamina breaksallow the invasion of epithelial cones into the subepithelial connectivetissue and the formation of microcholesteatomas. This mechanism may explain some types of human cholesteatomas, even those occurring behindan intact tympanic membrane.56 According to this theory, micro-cholesteatomas may enlarge and then perforate secondarily through thetympanic membrane, leaving the typical appearance of an attic chole-steatoma. This sequence of events has not been documented, althoughthe alternations in the differentiation of keratinocytes and basal celllayer of cholesteatoma matrix have been observed in several studies.

a small area in the anterior tympanum in the developing fetus oftencontains a small area of keratinizing epithelium. He found epidermoidformation in 37 of 68 temporal bones of fetuses at 10 to 33 weeksgestation. Congenital cholesteatomas may originate in this region.

Although congenital cholesteatomas generally are seen as pearl-likemasses behind an intact membrane, Koltai and coworkers29 presenteda convincing case that some congenital cholesteatomas advance toperforate and become chronically infected, taking on the appearance of an acquired cholesteatoma. Potsic and colleagues,30 in a review of 172series of congenital cholesteatomas, developed a useful staging system:stage I, limited to one quadrant; stage II, involving multiple quadrants

without ossicular involvement; stage III, ossicular involvement without

Figure 139-5. Cholesteatoma developing at the margin of perforation(secondary acquired cholesteatoma) with secondary infection.

Figure 139-6. Cholesteatoma behind intact tympanic membrane.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


5/16


immune responses, mechanisms of adaptive immunity also operate incholesteatoma.64 Data from Parisier and colleagues65 suggested that

broblasts in the subepithelium of cholesteatomas showed an invasivephenotype, whereas broblasts from postauricular and ear canal skinappeared either weakly invasive or not invasive. In a similar studyChole and colleagues66 found that normal broblasts and broblastsfrom induced cholesteatomas did not exhibit the invasive phenotypecharacteristics of true neoplastic cells.

Other lines of evidence support the basal cell hyperplasia/migration theory. Increased expression of human intercellular adhesionmolecule-1 and intercellular adhesion molecule-2 has been shown,suggesting a role in cell migration into tissue.40 The presence of heatshock proteins 60 and 70 suggested proliferation and active differen-tiation of basal keratinocytes in cholesteatomas.58 There are somereports that immune response is involved in the hyperproliferative stateof cholesteatoma epithelium.52,67,68 Langerhans cells may initiateimmune reaction and promote proliferation of keratinizing epitheliumvia an interleukin (IL)-1 and transforming growth factor (TGF)- mechanism.42,67,69

Abnormal distribution of epidermal differentiation markers, suchas laggrin and involucrin,57 c-jun and p53 proteins,58 and increasedepidermal growth factor receptor,59,60 has been shown in middle earcholesteatoma matrix. Increased levels of proteins cytokeratin 13 andcytokeratin 16, which are markers for differentiation and hyperprolif-eration, were also found.61 Kim and Chole39 showed the increasedexpression of cytokeratin 13 and cytokeratin 16 in the area of the periph-eral area of the pars tensa of induced cholesteatoma by ear canal ligationand in the peripheral and central area of pars tensa of induced cholest-eatoma by eustachian tube obstruction. Sakamoto and associates62 found ErbB-2 protein to be overexpressed, and cell proliferation andapoptosis of keratinocytes accelerated. Caspases play a key role in apop-tosis; Miyao and coworkers63 suggested that caspase-8, which is activatedby the induction of tumor necrosis factor- , leads to activation of caspase-3, which activates apoptotic nucleases in cholesteatoma tissue.

Human toll-like receptors are crucial in the induction and activa-tion of innate immunity in the course of an infection. Expression of toll-like receptor3 on the epithelium and some cells within the peri-matrix and the presence of T cells may suggest that apart from innate

Invagination

CholesteatomaBasal cell hyperplasia

Epithelial ingrowth

Squamous metaplasia

Retractionpocket

Epithelialcone

Perforation

Unknowncause

Secondaryperforation

Figure 139-7. Theories of pathogenesis of aural cholesteatoma.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


6/16


suspensory ligaments of the ossicles. The pars accida may invaginatinto the lateralmost portion of the epitympanum (Prussaks space) andthen into the recesses of the epitympanum posteriorly, lateral to thebody of the incus, inferiorly into the middle ear by way of the pouch

of von Trltsch (Fig. 139-8), or anteriorly into the protympanum(Fig. 139-9).73-75

c mpl aThe expansion of cholesteatoma may result in bone erosion of theossicles, otic capsule, fallopian canal, tegmen tympani, and tegmenmastoideum. These complications may cause intracranial complica-tions (Box 139-1). The erosion of ossicles, most commonly in the incus,may result in conductive hearing loss. The severity of hearing loss irelated to the status of the ossicles and the position of the cholesteatoma sac. Erosion of the otic capsule occurs most commonly in the lateralsemicircular canal and rarely in the cochlea. A labyrinthine stula hasbeen reported in 10% of cholesteatomas in adults and children,76 whichmay result in sensorineural hearing loss and vertigo. Sensorineuralhearing loss may result from the secondary suppurative labyrinthitis orfrom the cochlear hair cell loss adjacent to cholesteatoma.66 Facial nerveparalysis may occur acutely as a result of infection or slowly as a resulof expansion of the cholesteatoma. Erosion of the tegmen tympani orthe tegmen mastoideum may lead to development of a brain hernia orcerebrospinal uid leakage.77

Because cholesteatomas contain keratin debris enclosed in a tissuespace, they are subject to recurrent infection. The bacteria found ininfected cholesteatomas differ from bacteria found in AOM or OME.Signi cant anaerobic bacteria are present. The most common aerobicbacterium is Pseudomonas aeruginosa,and the most common anaerobicmicroorganism isBacteroides sp (Table 139-1).78

Ma ag mCongenital and acquired cholesteatomas can be eradicated from thetemporal bone only by surgical resection. The goals of surgery are to

Epithelial Invasion Theory The epithelial invasion theory 36 states that keratinizing squamous epi-thelium from the surface of the tympanic membrane invades or migratesinto the middle ear from a perforation in the tympanic membrane. Thistheory is supported by clinical observation and experimental evidence.

Weiss37 showed that epithelial cells could migrate along a surface by a process that he calledcontact guidance,and that, when they encounteranother epithelial surface, they stop migrating, for which he used theterm contact inhibition. van Blitterswijk and Grote38 reported thatcytokeratin 10, which was seen in the meatal epidermis and migrating epithelium, was preferentially expressed in the cholesteatoma matrix,rather than in the middle ear mucosa. This nding suggests an epider-mal origin of cholesteatoma. Kim and Chole39 showed increased cytok-eratin 10 expression in the peripheral area of the pars tensa of inducedcholesteatoma by ear canal ligation and in the peripheral and the centralarea of the pars tensa of induced cholesteatoma by eustachian tubeobstruction. The ndings of this study also support the basal cellhyperplasia hypotheses for the pathogenesis of aural cholesteatoma,

with regard to hyperproliferation, migration, and an altered differen-tiation of keratinocytes.39 High levels of bronectin and tenascin andfocal disruptions of the basement membrane, were reported in themiddle ear cholesteatoma, supporting the concept of the invasiontheory.40-43

This theory also is supported by investigations in animal choles-teatoma models and human temporal bones. Jackson and Lim44 gavehistologic and ultrastructural evidence that keratinizing epithelium canmigrate into the cat bulla by contact guidance. It is likely that in sometympanic membrane perforations, in ammation damages the innermucosal lining of the tympanic membrane, allowing the outer kerati-nizing epithelium to migrate inward and generate a cholesteatoma.Hueb and associates45 showed similar evidence in chinchillas. Palva andcolleagues46 showed histologic evidence for this theory in human tem-poral bones. Cholesteatomas originating after temporal bone fracturesmay result from this mechanism; fractures within the ear canal may allow ingrowth of keratinizing epithelium by contact guidance.47 7,12-Dimethylbenz[a]anthracene, a chemical carcinogen, could inducethe advancing of the keratinizing squamous epithelium into or underthe mucosal layer, and ingrowth and spreading over the middle earcavity and eustachian tube in the rat ear.48

Squamous Metaplasia Theor y Wendt70 theorized that the simple squamous or cuboidal epithelium of the middle ear cleft could undergo a metaplastic transformation intokeratinizing epithelium. Sad15,71 supported this theory, noting thatepithelial cells are pluripotent and can be stimulated by in ammationto become keratinizing. According to this theory, an area of keratiniz-ing epithelium within the middle ear would enlarge because of accu-mulated debris and contact with the tympanic membrane. Withintercurrent infection and in ammation, the cholesteatoma would leadto lysis of the tympanic membrane and perforation, resulting in thetypical appearance of an attic cholesteatoma. This theory is supportedby the demonstration that biopsy specimens from the middle ear of children with OME sometimes contain islands of keratinizing epithelium.15

Some experimental evidence supports the contention that middleear mucosa can become metaplastic and keratinize. Chole and Frush72 showed that extreme vitamin A de ciency leads to the formation of keratinizing epithelium within the middle ear and eustachian tube of rats. None of their experimental animals had developed cholesteatomas.Consequently, there is no direct evidence that cholesteatomas arise by squamous metaplasia of the middle ear mucosa.

From a clinical perspective, it seems that each of these pathogenicmechanisms accounts for a proportion of acquired cholesteatomas.Regardless of the pathogenesis of aural cholesteatomas, they all sharecertain properties. Cholesteatomas are prone to recurrent infection, andthey characteristically erode the bone of the ossicles and the otic capsule.

Aural cholesteatomas originating from the vicinity of the tympanicmembrane exhibit typical growth patterns into the temporal bone.Because most acquired cholesteatomas originate by invagination of the pars accida, their growth is limited by the mucosal folds and

Figure 139-8. Posterior mesotympanic cholesteatoma. This sac formsbecause of retraction of the posterior portion of the pars tensa, andfrequently invades the sinus tympani and facial recess. Extension tothe mastoid occurs medial to ossicle heads (arrows) . (From Jackler RK.The surgical anatomy of cholesteatoma. Otolaryngol Clin North Am . 1989;22:883.)
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


7/16


bene ts in some patients with only-hearing ears. Irrigation with 1 :1 : 1distilled white vinegar, distilled water, and 70% isopropyl alcohol maykeep some cholesteatomas stable if their opening into the ear canal is

suf ciently large (Box 139-2). If surgery is required for a cholesteatoma in an only-hearing ear, careful preoperative evaluation and operativeplanning, perioperative use of antibiotics and steroids, ossicular recon-struction, and postoperative care should be considered.81

Chronic Otitis Media without Cholesteatoma Acute or recurrent infection of the middle ear may result in a perma-nent perforation of the tympanic membrane. Ears with chronic perfo-rations without cholesteatoma may be chronically or intermittently infected. Three times as many operations were performed in the UnitedStates for this disease as were performed for cholesteatomas.25 Paparella and Kim81a reported that of 375 primary tympanomastoid operationsfor chronic mastoiditis, two thirds were performed in ears with granu-lation tissue and without cholesteatoma.

D agTympanic membrane perforation (Fig. 139-10) may result from AOM,chronic otitis media, or trauma (injury or surgery). In some instances,a dry, simple perforation results from a single episode of AOM (i.e.,necrotizing otitis media). Perforation of the tympanic membrane,especially involving the tympanic anulus, may allow ingrowth of thekeratinizing epithelium of the ear canal or tympanic membrane, leading to cholesteatoma. An ear with a simple perforation may becomeinfected because of contamination from the ear canal or because of asmoldering infection in the mastoid. A simple perforation commonly is seen as a low-frequency conductive hearing loss clinically. Thi

nding is supported by experimental perforation in rats.82 The tym-panic membrane velocity was found to be decreased in the low fre-quency in a small perforation and in the high and low frequencies ina large perforation.83

eradicate disease and manage complications and, secondarily, to recon-struct the middle ear. The decision whether to perform surgery dependson the nature and extent of disease, the existence of complications,mastoid pneumatization, eustachian tube function, hearing status of both ears, the reliability of the patient, and the experience and skill of the surgeon.77,79 Surgical approaches include atticotomy, simple mas-toidectomy, canal wall-up or canal wall-down procedures, radical mas-toidectomy, modi ed radical mastoidectomy, and Bondy procedure.

The open (canal wall-down) and the closed (intact canal withfacial recess) procedures have advantages and disadvantages (Table 139-2). The reported results of both procedures vary. Residual diseaseand recurrent disease occur in 11% to 27% and 5% to 13% of patientsundergoing the closed procedure, whereas residual or recurrent diseaseoccurs in 2% to 10% of patients undergoing the open procedure.80 Inthe cases of labyrinthine stula, facial nerve paralysis, and intracranialcomplications, surgery should be performed as soon as possible.

In some patients, a cholesteatoma can be dbrided of entrappedkeratin by direct removal or by irrigation. In some cases, surgical inter-vention is impossible or not advisable; the patient may not be medically able to withstand surgery, or the risks of surgery may not outweigh the

Figure 139-9. Anterior epitympanic cholesteatoma. Invagination of theepitympanum anterior to the malleus head and neck creates a choles-teatoma sac that threatens the horizontal facial nerve and geniculateganglion. Forward extension into the supratubal recess is common(arrows ). (From Jackler RK. The surgical anatomy of cholesteatoma. OtolaryngolClin North Am . 1989;22:883.)

Table 139-1

Bacteriology of Infected Cholesteatomasin 30 Children and Adults

Ba a n . ca

AerobesPseudomonas aeruginosa 11

Pseudomonas uorescens 2Streptococcus sp 8Proteus sp 4Escherichia coli 4Klebsiella, Enterobacter, Serratia 4

Alcaligenes, Achromobacter 3Staphylococcus epidermidis 2Staphylococcus aureus 1CBC group F

AnaerobesBacteroides sp 13Peptococcus, Peptostreptococcus 11Propionibacterium acnes 8

Fusobacterium sp 4Bi dobacterium sp 3Clostridium sp 3Eubacterium sp 2

Modi ed from Harker LA, Koontz FP. The bacteriology of cholesteatomas. In: McCabe BF, Sad J, Abramson M, eds.Cholesteatoma: First International Conference . New York:Aesculapius Publishers; 1977.

Box 139-1

Complications and Emergency States of ChronicOtitis Media with Cholesteatoma

Hearing lossconductive, sensorineural, mixed typeLabyrinthine stulamainly horizontal semicircular canal, rarely

cochleaFacial nerve paralysisacute or chronicIntracranial infectionsBrain hernia or cerebrospinal uid leakage
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/16


were only two constant openings: (1) between the tendon of the tensortympani muscle and the stapes, and (2) between the short process of the incus and the stapedial tendon. Edema and in ammation withgranulation tissue may block these communicating openings, prevent-ing drainage of the antrum and mastoid. Chronic obstruction of theattic and antrum with infection leads to irreversible changes in themucosa and bone of the antrum and mastoid. Granulation tissue withinthe temporal bone can lead to bone erosion; 4 of 123 temporal bonesin this series had active bone erosion. Thomsen and colleagues84 foundthat bone erosion in patients with chronic otitis media was moreprevalent when cholesteatoma was present, but it still occurred in theabsence of cholesteatoma. Moriyama and associates85 found osteoclasticbone resorption in rats with chronic otitis media.

Chronic otitis media may occur in patients who have indwelling tympanostomy tubes. Otorrhea may be a complication of tympanos-tomy tube insertion; it has been reported to occur in 9% to 34% of children undergoing this procedure.86 Chronic otorrhea that is resistantto therapy has been reported to occur in 5.5% of children with tubes.87

Table 139-2

Advantages and Disadvantages of Canal Wall-Up and Canal Wall-Down Procedures inChronic Otitis Media with Cholesteatoma

Adva ag D adva ag

Canal Wall-UpPhysiologic position of tympanic membrane Residual and recurrent cholesteatoma may occurEnough middle ear space Incomplete exteriorization of facial recessNo mastoid cavity problem Second-stage operation often required

Canal Wall-DownResidual cholesteatoma easily found on follow-up evaluation Mastoid cavity problem oftenRecurrent cholesteatoma rare Middle ear shallow and dif cult to reconstructTotal exteriorization of facial recess Position of pinna may be altered; second-stage operation sometimes

required

Modi ed from Harker LA, Koontz FP. The bacter iology of cholesteatomas. In: McCabe BF, Sad J, Abramson M, eds. Cholesteatoma: FirstInternational Conference . New York: Aesculapius Publishers; 1977.

Figure 139-10. Tympanic membrane perforation.

Box 139-2

Management of Chronic Otitis Mediawith Cholesteatoma

Factors for Determination of TreatmentExtent of diseasePresence of complicationsHearing states of both earsEustachian tube functionMastoid pneumatizationPatient factors: general medical condition, age, occupation,

reliability Surgeons skill

Conservative TreatmentRemoval of entrapped keratin: directIrrigation with 1 : 1: 1 distilled white vinegar, distilled water, and

70% isopropyl alcohol for stabilization

Surgical ApproachAtticotomy: transcanalSimple mastoidectomy Canal wall-up procedure (intact wall) with or without facial

recess approachCanal wall-down procedure (canal down): radical or modi ed

radical mastoidectomy, Bondy procedure

Pa h gChronic otomastoiditis without cholesteatoma is marked by the pres-ence of irreversible in ammatory changes within the middle ear andmastoid. The factors that allow acute infections within the middle earand mastoid to develop into chronic infections are unclear. da Costa and colleagues27 found granulation tissue in 96%, ossicular changes in96%, tympanosclerosis in 43%, cholesteatoma in 36%, and cholesterolgranuloma in 21% of the human temporal bones of patients withchronic otitis media with a perforated tympanic membrane (Table 139-3). Aeration of the middle ear, antrum, and mastoid depends onthe free movement of gases from the eustachian tube into the mastoidair cells. In the human temporal bone, gases must travel around theossicles in the epitympanic space to get into the antrum (Fig. 139-11).

Proctor74 showed that the middle ear is separated from the antrumnot only by the ossicles, but also by mucosal folds. He found that there
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


9/16


toxicity in rodents91 and primates.92 Although there are reports suggest-ing that sensorineural hearing loss may occur after topical use of thesepreparations,93,94 no conclusive evidence is available proving ototoxicityof commercially available otic preparations in the human middle ear.

Potentially ototoxic topical preparations should be applied to themiddle ear only when potential bene ts outweigh the potential risks.Topical antibiotics also can be applied in powder form by insuf ation.Various agents have been used individually or in combination, includ-ing boric acid, sulfamethoxazole, chloramphenicol, and hydrocortisone.This technique is particularly useful in the presence of epithelialitis andin a moist mastoid cavity, although patients may become sensitized tosulfamethoxazole or amphotericin. Systemic antibiotics should be usedin refractory cases when speci c pathogens are found on culture. Severaquinolones such as cipro oxacin, o oxacin, and nor oxacin may beuseful in these patients. In ears that become repeatedly infected butclear between episodes, tympanoplasty should be considered. Ideallyan ear with a tympanic membrane perforation should be free frominfection for 3 months before tympanoplasty.

In some patients, chronic infection with otorrhea, but withoutcholesteatoma, persists despite aggressive medical therapy. In these

cases, two options should be considered: long-term (6 to 8 weeks)intravenous antibiotics or tympanomastoid surgery. Antibiotic therapy can be administered at home or in the hospital. Aggressive local d-bridement is required. The goals of tympanomastoidectomy includeaeration of the middle ear and mastoid, removal of irreversibly diseasedtissue, closure of the middle ear, and reconstruction of the sound-conducting mechanism. These goals are not always achieved in a singlestage.

Otorrhea occurs after insertion of tympanostomy tubes in approx-imately 10% of ears. The incidence of immediate postoperative otor-rhea usually can be decreased by the use of topical antibiotics at thetime of tube insertion.95,96 The use of Silastic tympanostomy tubesimpregnated with silver oxide was shown to decrease the incidence ofotorrhea signi cantly in a randomized clinical trial.86 If infection per-sists, intravenous antibiotic therapy directed at speci cally identi edorganisms should be considered. Because chronically infected tym-panostomy tubes may have persistent bacterial bio lms, removal of thetympanostomy tube may be necessary in some refractory cases.90,97

Bone Erosion in Cholesteatoma andChronic Otitis MediaIn Virchows description of the pathology of aural cholesteatoma in1854, he noted that the cholesteatoma extended through the bone tothe external auditory canal, sometimes, also, in the cranial cavity.98 Since that time, clinicians and investigators have studied the patho-physiology of bone resorption from this disease. Although muchprogress has been made in the understanding of the resorptive process,the actual sequence of events and their relative importance are notcompletely understood.

It is unclear whether such chronic infection is a result of the indwelling tube or of the drainage of an already smoldering infection. Giebink andcoworkers88 found the risk of otorrhea on the second postoperative day

was signi cantly increased by the presence of a bacterial pathogen inthe ear canal or in the middle ear effusion, and by the presence of in amed middle ear mucosa at tympanostomy tube insertion. Thebacteria that are found in such cases are bacteria found in chronic otitismedia generally: P. aeruginosa and Staphylococcus aureus . Erkan andcolleagues89 reported the culture results of aspiration of exudate in 183patients with chronic otitis media. Aerobes were found in 39% of thesamples, anaerobes were found in 11%, and aerobes and anaerobes werefound in 50%. P. aeruginosa, S. aureus,and Klebsiella pneumoniae werecommonly found in aerobes, and Bacteroides sp were most commonly isolated in anaerobes. The colonization of tympanostomy tubes in theform of microbial bio lms may lead to persistence and recurrence of infection.90

Ma ag mMost infected perforations can be managed conservatively with topicalantibiotics. Antibiotic otic suspensions with or without steroids usually are effective. The antibiotics should be chosen to eradicate the mostcommon pathogens, P. aeruginosa and S. aureus . In cases with recurrentor chronic infections, cultures should be used to adjust antibiotics.Many topical otic antibiotic preparations contain potentially ototoxicsubstances, including aminoglycoside antibiotics and propylene glycol.Studies of these substances applied to the middle ear have shown oto-

Table 139-3

Pathologic Findings in the Temporal Bones with Chronic Otitis Media

Pa h l g f d g P a d tympa M mb a ( n = 116) (%) n p a d tympa M mb a ( n = 28) (%)

Granulation tissue 113 (97.4) 27 (96.4)

Ossicular changes 105 (90.5) 27 (96.4)

Tympanosclerosis 23 (19.8) 12 (42.9)

Cholesterol granuloma 14 (12.1) 6 (21.4)

Cholesteatoma 5 (4.3) 10 (35.7)

Modi ed from da Costa SS, Paparella MM, Schachern PA, et al. Temporal bone histopathology in chronically infected ears with intact andperforated tympanic membranes. Laryngoscope . 1992;102:1229.

AntrumAttic

Chordatympani

nerve

Eustachian tubeorifice

Roundwindowniche

Mastoid

Figure 139-11. Aeration of the antrum and mastoid depends on com-munication around the ossicle within the attic. Obstruction of thesenarrow openings may result in chronic infection.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


10/16


and coworkers118 indicated an increased number of osteoclast precursorcells in the perimatrix of cholesteatoma tissue concomitant with anenhanced expression of RANKL, osteoprotegerin, and macrophagecolony-stimulating factor. The cholesteatoma-induced in ammatory process is associated with the release of RANKL from stromal cells anactivated T cells, which triggers osteoclastogenesis.113,118

The localized control of osteoclastogenesis is complex, and signaling pathways may vary during the natural course of an in ammatoryprocess. The type of bone undergoing resorption may also modify theprocess. The endochondral bone of the otic capsule seems to be moreresistant to erosion than the intramembranous bone of the middle earand mastoid.119

Some investigators have noted that the keratin debris within a cholesteatoma sometimes extrudes into the subepithelium adjacent tobone eliciting an in ammatory and osteoclastic response. Yuasa andcolleagues120 showed that the pH of keratin debris within cholesteato-

mas was acidic, which they believed led to demineralization of thehydroxyapatite of bone. Some investigators have shown that the keratinitself may induce an in ammatory reaction (foreign body granuloma),

which leads to cellular bone resorption.85,121,122The extracellular matrix exhibits alterations in cholesteatoma, and

many studies have evaluated the role of enzymes in cholesteatoma-induced bone resorption. By immunocytochemistry and zymography,the expression of matrix metalloproteinases (MMP-1, MMP-2,MMP-3, and MMP-9) was found to be strictly con ned to the basaland suprabasal cell layers of the cholesteatoma epithelium.123-126 Theneutrophil collagenase showed a more disseminated expression in theepithelium and the granulation tissue. The tissue inhibitor of metal-loproteinases, TIMP-1, could be detected only in very limited areas of the granulation tissue in a quite randomized manner.126 Because of theirdestructive capacity, matrix metalloproteinases are normally tightlycontrolled. A derailment of these regulatory mechanisms in favor ofproteolysis was postulated to play a role in the invasion of cholesteato-mas into the temporal bone.125,126

Collagenase plays an important role in the mechanism of localinvasion by aural cholesteatoma.116,127 Neutral collagenase may stimu-late osteoclastic resorption by degrading the osteoid surface of bone,allowing osteoclastic activity.128 Neutral collagenase has not been found

within osteoclasts,129 but has been localized in the vicinity of resorbing bone.116 In the collagenase cascade, procollagenase is activated to col-lagenase by plasmin. Plasminogen is activated to plasmin by plasmino-gen activator. Plasminogen activator is thought to play a role inkeratinocyte proliferation, migration, and differentiation.130 Plasmino-gen activator can be divided into two types: tissue-type plasminogenactivator and urokinase-type plasminogen activator. Tissue-type plas-minogen activator is known to activate osteoclasts in vitro.131 Naka-

In the rst half of the 20th century, it was believed that the boneresorption seen adjacent to cholesteatomas was a result of pressurenecrosis,99 although clinical observations led to the abandonment of thepressure necrosis theory. It was thought to be highly unlikely thatcholesteatomas could exert pressure exceeding capillary perfusion pres-sure (approximately 25 mm Hg), and this was later con rmed by Orisek and Chole100 using direct measurements of the pressure exerted by experimental cholesteatomas (1.3 to 11.9 mm Hg). Lautenschlager101

rst suggested that cholesteatomas may elaborate enzymes that directly erode bone, and some investigators showed that cultured human cho-lesteatoma could degrade guinea pig collagen.63,102 Direct extracellulardissolution of bone was never shown, however.

It is now clear that in ammatory processes in the temporal boneinduce the development and activation of osteoclasts, the only cellcapable of resorbing bone. Several early human temporal bone studiesrevealed osteoclastic bone resorption adjacent to cholesteatoma

matrix.103,104

Because of the relative scarcity of osteoclasts seen in surgi-cal specimens, however, numerous investigators hypothesized thatmononuclear histiocytes were the predominant cell associated witherosive cholesteatomas.105-107 Bernstein and colleagues108,109 noted thatearly studies were based on biopsy material taken during tympanomas-toid surgery, and surgery often is preceded by attempts to control activein ammation; biopsy material may represent the reparative phaseinstead of the resorptive phase. Chole110 showed ultrastructural evi-dence in human and experimental cholesteatomas that bone resorptionis primarily a result of the action of multinucleated osteoclasts on bone(Figs. 139-12 and 139-13). Although many mononuclear cells (histio-cytes and broblasts) were present in the vicinity of active bone resorp-tion, only multinucleated osteoclasts were seen to disrupt the lamina limitans of bone and cause resorption lacunae. It is established thatmultinucleated osteoclasts caused bone resorption in patients withcholesteatoma and chronic otitis media.110-114

For bone resorption to occur, enzymatic removal of the organicand inorganic components must occur. These enzymes likely areelaborated or activated by the resorbing cells (osteoclasts) in theirimmediate microenvironment. These enzymes include acid phos-phatase,54,110,115,116 collagenase,54,116 and acid proteases.117 In studies by Blair and colleagues,117 a cathepsin-like proteolytic enzyme withmaximal activity at pH 4.0 was shown to be active in the microenviron-ment of the ruf ed border of the osteoclast.

Numerous factors may lead to localized osteoclast recruitment andactivation in regions of in ammatory osteolysis. It has been shown thatbone remodeling and bone loss are controlled by a balance between thereceptor activator of NF- B (RANK) and its ligand, RANKL.113 TheRANKL receptor, or RANK, has been identi ed on dendritic cells,chondrocytes, osteoclast precursors, and mature osteoclasts.113 Hamzei

A

OsteoclastsOsteoclasts

Bone

0.5 m0.5 m

BoneBone

V

VV

RBRB

MIMI

BB

Figure 139-12. A, Multinucleated osteoclasts actively absorbing bone beneath human cholesteatoma in a surgical biopsy specimen. B, Transmis-sion electron micrograph of the ruf ed border (RB) of an osteoclast showing mitochondria (MI) in the cytoplasm and vacuoles (V) containing acidproteases in the eroding zone.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


11/16


Although the mechanism is unknown, osteoclast activity can beinhibited by bisphosphonates, also known as antiosteolytic agents suchas 1-hydroxyethylidene-1,1-bisphosphonate (HEBP), risedronate, andzoledronate. These drugs inhibit the localized recruitment and activa-tion of osteoclasts in vivo and in vitro.143,144

Proin ammatory cytokines play an important role in middle ear

infections and cholesteatoma. They are released by macrophages, Tlymphocytes, monocytes, and many other cells at the site of infection.IL-1 was identi ed in cholesteatoma matrix,8 and was shown to stimu-late broblasts and macrophages to produce PGE2 and collagenase.105 Cultured cholesteatoma specimens produced IL-1 and IL-1, andboth isoforms have been detected within cholesteatoma tissue.41,145,146 IL-1 and IL-1 are potent inducers of bone resorption and may actby increasing PGE2. Human recombinant IL-1 receptor antagonistblocked bone resorption in mouse calvaria as a result of IL-1, but notIL-1 .147

Concomitantly, a parallel process of osteoblastic bone depositioninvariably accompanies osteoclastic bone resorption. Factors known toinduce new bone formation such as transforming growth factor (TGF)- and bone morphogenetic protein-2 have been identi ed in choles-teatomas.105,148,149 Cytokines such as TGF-1 and TGF-2 may potentially slow the proliferation and tissue destruction associated withhuman cholesteatoma.150 TGF- is involved in matrix formation andstimulates numerous matrix proteins, such as collagen, laminin, and

bronectin. TGF- may also initiate bone formation by recruitmentand proliferation of osteoblast precursor, whereas bone morphogeneticprotein-2 seems to be important in inducing differentiation of pluripo-tent progenitor cells.

Fujioka and Huang 151 identi ed platelet-derived growth factor inhuman cholesteatoma tissue, which stimulates monocytes to formmultinucleated osteoclast-like cells. Additionally, Bujia and coworkers60 and Sudhoff and colleagues59 showed abnormal expression of epidermalgrowth factor in human cholesteatoma, which is a potent stimulator of cell proliferation and differentiation. Production of parathyroid hor-monerelated protein was detected in cholesteatoma cell cultures andmay increase incidence of bone resorption.152

mura and associates132 showed both types of plasminogen activator intwo of six tissue extracts of human cholesteatoma.

Ohsaki and colleagues133 showed that bone adjacent to choles-teatomas is demineralized. Although their study suggested that dem-ineralization by acidic cholesteatomas may explain bony destruction,intimate contact of the cholesteatoma matrix with bone does not seem

to be necessary for bone resorption. Bone resorption has been shownto occur in patients with chronic otitis media with or without choles-teatoma.84 Macri and Chole134 showed that an implanted siliconebarrier between a cholesteatoma and underlying bone did not preventosteoclastic bone resorption in experimental cholesteatoma. It is likely that indirect effects (e.g., pressure or in ammation) may activate thecellular events of bone resorption.135 More recent studies have shownthat pressure with85,136 or without32,137 in ammation is suf cient toinduce bone resorption in experimental animals. The histologic appear-ance of pressure-induced bone resorption is similar to that seen incholesteatoma, showing osteoclasts and granulation tissue.111 It hasbeen postulated that the physical effects (pressure) of cholesteatomasmay lead to transient electrical potentials138 and the recruitment of monocytes into the subepithelial space. These monocytes may activatethe downstream cellular events leading to bone resorption.

Activated monocytes can produce arachidonic acid metabolitessuch as prostaglandins and leukotrienes, which are potent in ammatory mediators. Prostaglandin E2 (PGE2)139 has been shown to stimulatebone remodeling similar to the kallikrein-kinin system.51 Prostaglan-dins, produced from the cyclooxygenase pathway of arachidonic acidmetabolism, can be inhibited by aspirin, indomethacin, and ibuprofen.Indomethacin has been shown to impede bone resorption by inhibiting the production of PGE2 in vitro.140 This study also showed that in vivoindomethacin can inhibit pressure-induced bone resorption in ananimal model. Ibuprofen, another prostaglandin inhibitor, has alsobeen shown to inhibit bone resorption in a dose-dependent manner.141 Leukotrienes are 5-lipoxygenase metabolites of arachidonic acid.

Although the role of leukotrienes in cholesteatoma is unclear, the pep-tidoleukotrienes (LTC4, LTD 4, and LTE 4) stimulate isolated osteoclaststo accumulate tartrate-resistant acid phosphatase and resorb bone.142

Bone

Resorption

Biofilm

LPSLPS

CholesteatomaCholesteatoma

InflammationT Cells

Macrophages

Proinflammatory CytokinesTNF, IL-1, IL-6

RANKL

RANKL

Bone Marrow Monocytes

RANK

RANK

RANK

RANK

R e s t i n g O s t

e o b l a s t s

Figure 139-13. The bone resorption associated with cholesteatomas occurs as a result of the recruitment of osteoclasts. Osteoclasts are derivedfrom bone marrow mononuclear cells. The receptor activator of NF- B ligand (RANKL) on osteoblasts or T cells binds to receptor activator of NF- B (RANK) on monocytes forming osteoblasts. Osteoclast activation is potentiated further by numerous proin ammatory cytokines (tumornecrosis factor [TNF], IL [interleukin]-1, IL-6) that are produced in response to factors produced by the cholesteatoma or bacterial products suchas lipopolysaccharide (LPS).
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


12/16


Figure 139-14. Tympanic membrane tympanosclerosis (myringo-sclerosis).

Vascularization within the perimatrix of cholesteatoma showed a vefold increase compared with middle ear mucosa and a twofold

increase compared with skin.152 Sudhoff and colleagues153 found a closerelationship between the density of capillaries, degree of in ammation,and expression of different angiogenic factors. Because proliferating tissues such as middle ear cholesteatoma require an enhanced bloodsupply, angiogenesis seems to be necessary for the expansion of cho-lesteatoma matrix within the middle ear cavity.

The role of other cells in the pathophysiology of bone erosion inpatients with chronic otitis media is unclear. Gantz154 suggested thatLangerhans cells within the matrix of cholesteatomas might initiate animmunologic response to the presence of antigens (keratin and bacterialdebris); these cells may induce the cellular events outlined previously.These cells may also promote proliferation of keratinized epithelial cellsby means of IL-1 in vitro.92 Aberg and associates155 found no increasein Langerhans cells compared with ear canal epithelium, however, sug-gesting that these cells may not have a primary role in the developmentof cholesteatoma. Mast cells are seen in cholesteatoma matrix, but theirrole is unknown.156

The important role of infection in the pathogenesis of chole-steatoma is highlighted by the more recent identi cation of bacterialbio lms in otitis media 157 and cholesteatoma.35,158 Bio lms are bacterialcommunities enclosed in a self-produced matrix adherent to a surface.147 Many bacterial species relevant to otologic infections are known toform bio lms, including P. aeruginosa, H. infuenzae, Streptococcus

pneumoniae,and S. aureus .34 Bacteria within bio lms are more resistantto antibiotics and are likely responsible for the chronicity and recur-rence of these infections. The presence of antibiotic-resistant bacterialbio lms in cholesteatomas may also explain their aggressiveness.35 Bac-terial bio lms within cholesteatomas may elaborate lipopolysaccharideand other bacterial products that stimulate osteoclastogenesis. Zhuang and colleagues159 showed that lipopolysaccharide derived fromP. aeru-

ginosa can induce osteoclast development in vitro and potently stimu-lates in vivo bone resorption through a toll-like receptor4dependentmechanism.159

Sensorineural Hearing LossThe destructive effects of an expanding cholesteatoma within themiddle ear or mastoid and concomitant chronic infection are notlimited to the bony structures of the temporal bone. Paparella and

coworkers160

observed sensorineural hearing loss in patients withchronic otitis media, and Chole and Chiu161 observed loss of cochlearhair cell stereocilia in animals with experimental cholesteatomas withor without infection. In a study of age-matched, staged cholesteatomas,McGinn and Chole 162 showed loss of cochlear hair cells in areas subja-cent to areas of bone erosion, suggesting that ototoxic substances may traverse the bony wall of the cochlea directly. Meyerhoff and associ-ates163 found that 17.9% of temporal bones with chronic otitis media had histologic evidence of labyrinthitis.

Vartiainen and Karjalainen164 compared 874 chronically infectedears with and without cholesteatoma with 609 control ears, and foundsigni cantly worse bone conduction in the infected group; ears withcholesteatomas were generally worse than ears without cholesteatomas.Fria and coworkers165 found that there was a tendency for bone conduc-tion thresholds to be elevated at 2 kHz and 4 kHz in the ears withchronic OME compared with bone conduction thresholds in children

with or without OME. In animal experiments, the inner ear is particu-larly sensitive to injury by middle ear infection. Morizono and associ-ates166 found that otitis media in chinchillas resulted in signi cantincrease in tone-burst elicited compound action potentials after theotitis had cleared, indicating a sensory hearing loss.

In contrast, after correcting for the arti cial elevation of boneconduction thresholds from conductive hearing loss, also known as theCarhart effect, Browning and Gatehouse167 found no difference in thebone conduction thresholds between 395 ears with chronic otitis media and 920 control ears. These results were supported by MacAndie andOReilly,168 who pointed out that the presence of cholesteatoma orossicular erosion, or both, was not associated with a signi cantly increased risk of sensorineural hearing loss. Rahko and colleagues169 studied bone conduction thresholds in 359 children with a history of

recurrent AOM and found no correlation between the number of boutsof infection and permanent sensorineural hearing loss.

Tympanosclerosis

D agTympanosclerosis is thought to be a complication of otitis media in

which acellular hyalin and calci ed deposits accumulate within thetympanic membrane and the submucosa of the middle ear. In mostpatients, these plaques are clinically insigni cant and cause little or no

hearing impairment. Tympanosclerotic plaques within the tympanicmembrane appear as a semicircular crescent or horseshoe-shaped whiteplaque within the tympanic membrane (Fig. 139-14).

Pa h gTympanosclerosis is a consequence of resolved otitis media or trauma.Hussl and Mueller170 found tympanosclerosis to be a frequent sequela of chronic OME; they found tympanosclerosis in 19.7% of drumheads6 to 8 years after the insertion of ventilating tubes for OME. They alsonoted that middle ear tympanosclerosis often was seen after recurrentbouts of AOM. Tos and Stangerup171 found a signi cant increase intympanosclerosis in ears in which grommets were placed (59%) com-pared with the contralateral ears, in which only myringotomy wasperformed (13%). Daly 172 reported the weighted average incidence of tympanosclerosis is 10% in children 4 to 15 years old, with an averagefollow-up period of 4 years. The incidence of tympanosclerosis inchronic otitis media has been reported to range from 9% to 38%.Kinney 173 found that 20% of 1495 patients undergoing surgery forchronic otitis media or its sequelae had tympanosclerosis. Mangat andcoworkers174 found that 23.6% of 1274 patients treated with tym-panostomy tubes had tympanosclerosis.

Tympanosclerosis appears histologically as an acellular hyalinization of the subepithelial connective tissue of the tympanic membraneand middle ear. In most instances, calci cation is present. Osteoneo-genesis also can occur within these lesions. The bone deposition andossicular xation occur most frequently in the attic associated with theheads of the malleus and incus. When plaques occur within the tym-panic membrane, they are limited to the lamina propria. Hussl andLim175 found these plaques to be a degenerative process resulting incalci cation in connective tissue of the middle ear. They hypothesized
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


13/16


PetrositisInfection of the mastoid and middle ear may be complicated by thespread of infection within the temporal bone into the petrous apex.Petrous apicitis is an extension of infection from the mastoid air celltract into a pneumatized anterior or posterior petrous apex. In thepreantibiotic era, otitis media often was complicated by a spread intothe petrous apex and further intracranial complications. The classicsymptoms of petrous apicitis include deep facial pain, otitis media, andipsilateral abducens nerve paralysis. This triad, called Gradenigos syndrome,183 is rare, although suppurative processes in the petrous apex

occur in patients with AOM and chronic otitis media, but most oftenmanifest as chronic infection with otorrhea and sometimes deep painafter adequate surgery. More recently, otologic diseases have beenobserved as complications in patients with acquired immunode ciencysyndrome (AIDS). Chandrasekhar and colleagues184 reported six of seven specimens had petromastoiditis in the temporal bones of AIDSpatients.

H al nThe history of petrous apicitis and its management has been reviewedmore recently.185 A patient with petrous apicitis and Gradenigos triad

was rst described by Goris.186 In Gradenigos review of previously published cases of 57 patients with petrositis, 24 actually had the puretriad; others had multiple complications. In the early part of the 20thcentury, there was a controversy as to whether petrous apicitis can developin diploic (marrow- lled) or pneumatic (air- lled) petrous apices.187 Itis generally believed that petrous apicitis occurs in patients who havepneumatized petrous apices. In the 1930s, Almour188 and Kopetzky and

Almour189 described surgical approaches for petrous apicitis in whichstulous tracts were followed into the petrous apex. Ramandier in

1933,190 and soon thereafter Lempert191 described the now classic oper-ation for exenteration of the anterior petrous apex. The histopathology of petrous apicitis was described by Lindsay.192 An additional surgicalapproach for suppuration of the petrous apex was described in a casereport by Hendershot and Wood, 193 in which they drained an osteo-myelitis of the petrous apex through the middle cranial fossa.

A a myThe petrous apex is a truncated pyramid, which is the portion of thetemporal bone medial to the inner ear labyrinth (Fig. 139-17). The

that OME or AOM led to a destructive process within connectivetissue, which led to degeneration of collagen and subsequently dys-trophic calci cation and tympanosclerosis.

The degeneration of collagen may be a direct result of in amma-tion or infection within the middle ear (e.g., by bacterial proteinasesand collagenases). Wielinga and colleagues176 showed that eustachiantube obstruction alone, without infection, caused tympanosclerosis inrats; they hypothesized that deformation alone was suf cient to causethe plaques to form. Another possible cause of tympanosclerosis is anautoimmune process occurring within the tympanic membrane. Schiff and associates61 prepared antisera to guinea pig lamina propria andpassively immunized guinea pigs. When the tympanic membranes of

these animals were traumatized, tympanosclerotic plaques developed.Chole and Henry 177 found that LP/J inbred mice spontaneously devel-oped middle ear lesions that resembled tympanosclerosis and may beimmunologically mediated.178 Hussl and Lim175 proposed two possiblemechanisms for the formation of tympanosclerotic plaques, beginning

with collagen degeneration (Fig. 139-15). Russell and Giles179 foundthat the process of tympanosclerosis started in the submucosal connec-tive tissue layer and progressed to involve all connective tissue sublayersin an animal model. The extent of calcium deposition and brosisacross the membrane was related to the duration of OME.

Ma ag mTympanosclerosis within the middle ear (Fig. 139-16) is histologically similar to tympanosclerosis occurring within the tympanic membrane,but it often leads to conductive hearing loss caused by ossicular xation.

Although some authors have stated that tympanosclerosis tends to recurafter surgical removal, others have reported stable hearing results inthese patients. Smyth and colleagues180 reported excellent hearing results in 79% of tympanosclerotic ears in which ossicular reconstruc-tion (stapedectomy and total ossicular reconstruction) was performedin two stages, although Gormley 181 found that only 7% of his cases hadan air-bone gap of less than 21 dB on long-term follow-up evaluation,questioning the advisability of stapedectomy in ears with tympanoscle-rosis. In the earlier series182 in which one-stage procedures were per-formed, 21% of 57 cases resulted in cochlear losses. Tympanoplasty and ossicular reconstruction can be performed in ears with tympano-sclerosis, but the risks of cochlear damage seem to be greater than inother middle ear diseases because of the extensive dissection that isrequired in tympanosclerotic ears and the coexistence of labyrinthineerosion.

Connective tissue degeneration

Calcified tympanosclerotic plaques(ossification)

Various pathogenic factors(such as inflammation, autoimmunity, trauma)

Fibrolysis

Hyalinization(hypovascularity)

Change in pH

Ca-phosphate precipitates

Dystrophic calcification

Fibrocyte degeneration

Extracellular matrix vesicleswith Ca and PO

4ions

Supersaturation

Ca-phosphate precipitates

Matrix vesicle calcification(calcospherules)

Figure 139-15. Two possible mechanisms for formation of tympano-sclerotic plaques. (Adapted from Hussl B, Lim DJ. In: Lim DJ, Bluestone CD,Klein JO, eds. Recent Advances in Otitis Media with Effusion . St. Louis: Mosby; 1984.)

Figure 139-16. Advanced tympanosclerosis of the tympanic mem-brane and the middle ear.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


14/16


the cranial nerve; 6 (27.3%) had facial nerve paralysis, 4 (18.2%) hadabducens nerve paralysis, and 2 had coma (Table 139-4).197 Patients

with suppuration may manifest various symptoms, none of which arepathognomonic for the syndrome. In patients with long-standing chronic otomastoiditis, deep pain, and persistent infection, the diagno-sis of petrositis should be considered. In a more recent series of patients

with petrous apicitis, the predominant organism wasP. aeruginosa . Thephysical ndings of petrositis usually include those of chronic otitismedia with chronic otorrhea. In some patients, the infection can be

petrous apex is the most surgically inaccessible portion of the temporalbone.194 The apex may be arbitrarily bisected by a coronal plane throughthe internal auditory canal (seeFig. 139-17). This plane divides theapex into an anterior portion, the peritubal area, and a posteriorportion, the perilabyrinthine area. The posterior petrous apex, whichis pneumatized in 33% of patients, is just medial to the semicircularcanals. The anterior apex, which is pneumatized in 10% of patients, isanterior and medial to the cochlea. The carotid artery traverses theanterior petrous apex.

The petrous apex may be pneumatic (air cell lled), diploic(marrow lled), or sclerotic (solid bone). Direct extension of infectionfrom the mastoid and middle ear through pneumatized air cell tractsinto the petrous apex is thought to be the etiology of petrous apicitis.It has been estimated that 30% of posterior petrous apices are pneu-matized, and in a study of 84 normal human temporal bones, 9% of anterior petrous apices were found to be pneumatized (Fig. 139-18).195

The anatomic relationship at the petrous tip may explain some of the symptoms of petrous apicitis. An undetected and poorly drainedinfected air cell of the petrous apex must trail through small air celltracts into the middle ear and mastoid. These cell tracts consist of theinfralabyrinthine air cell tract, the retrofacial tract, and the peritubal aircells superior to the eustachian tube. If the bony cortex of the anteriorpetrous apex is involved by the extension of infection, the infectionmay cause an epidural abscess in the region or damage nearby cranialnerves. On the superior aspect of the petrous tip lies the trigeminal orgasserian ganglion. Damage or irritation to the ganglion may explainthe deep facial pain that some patients with apicitis experience. Extend-ing from the tip of the petrous apex to the clinoid is the petroclinoidligament. The abducens nerve travels below the petroclinoid ligamentin a small canal called Dorellos canal.196 Entrapment or in ammationin the ear of Dorellos canal is thought to account for the presence of abducens paralysis in some patients with petrous apicitis.

D agSymptoms of petrositis usually are subtle. Typically, a patient who hashad previous mastoid surgery complains of persistent infection anddeep facial pain. In a series of 22 patients during 20 years, 16 patients(72.7%) had otalgia, and 13 patients (59.1%) had deep facial pain andheadache; trigeminal nerve paralysis (68.2%) most frequently involved

APA

PPA

IPSIPS

FM

FL

FOFS

Figure 139-17. Base of skull viewed from above, revealing the relationship of the petrous apex to the foramen magnum (FM) and the rest of thetemporal bone ( box, enlarged on right). The anterior petrous apex (APA) is a truncated pyramid based on labyrinth (shaded) and bound by foramenlacerum (FL) anteriorly and inferiorly and inferior petrosal sinus (IPS) posteriorly. If the petrous apex is viewed from above, it can be divided intoanterior and posterior sections by a line drawn through the internal auditory canal. APA is located medial to the cochlea and internal auditory canal, and posterior petrous apex (PPA) is medial to the semicircular canals. FO, foramen ovale; FS, foramen spinosum. (Modi ed from Chole RA.Petrous apicitis: surgical anatomy. Ann Otol Rhinol Laryngol . 1985;94:251.)

Table 139-4

Symptoms Found in 22 Patients withPetrous Apicitis from 1976-1995

symp m n . Pa (%)

Deep pain and headache 13 (59)

Otalgia 16 (72)

Otorrhea 13 (59)

Fever 5 (22)

Coma 2 (9)

Cranial nerve paralysisV 15 (68.2)VI 4 (18.2)VII 6 (27.3)VIII 9 (40.9)IX 1 (4.5)X 1 (4.5)

From Chole RA, Gadre AK. Petrous apicitissymptomatology,pathology, and management [Abstract]. Skull Base Surgery Symposium, Sacramento, CA, 1995.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


15/16


to control the suppuration, various surgical approaches are available.Surgical therapy aims to achieve drainage of the petrous apex throughthe mastoid and middle ear into the petrous apex. These air cells havebeen well de ned anatomically 194; they include the subarcuate andsinodural angle cells toward the posterior petrous apex and the peritu-bal, retrofacial, infralabyrinthine, and infracochlear tracts toward theanterior petrous apex. The anterior apex may be widely exposed throughthe glenoid fossa with the approach of Ramandier190 and Lempert.191 If adequate air cells cannot be identi ed through the middle ear andmastoid, the middle cranial fossa approach can be used to enter theroof of the anterior petrous apex.193 Brackmann and Toh138 found thetranslabyrinthine approach useful in nonhearing ears. In hearing indi-viduals, anatomy permitting, the transcanal infracochlear approach

with stenting was the preferred approach for drainage of the petrousapex.

SummaryOtitis media is one of the most common diseases of childhood, and themost frequent disease managed with antibiotics in children. The infec-tious and noninfectious complications of otitis media may result insigni cant morbidity and complications, including acute and chronicmastoiditis, petrositis, and intracranial infection. The noninfectioussequelae, including chronic perforation of the tympanic membrane,ossicular erosion, labyrinthine erosion, and tympanosclerosis, are majorcauses of hearing loss.

Acute or recurrent infection of the middle ear may result in a permanent perforation of the tympanic membrane and irreversiblein ammatory changes within the middle ear and mastoid known aschronic otomastoiditis. Chronic otitis media may occur in patients whohave indwelling tympanostomy tubes. Most cases of chronic otitismedia can be managed with topical antibiotics.

Aural cholesteatomas are epidermal inclusion cysts of the middleear or mastoid, and are classi ed as congenital and acquired. Acquiredcholesteatomas possibly result from OME or AOM or both. There arefour basic theories of the pathogenesis of acquired aural cholesteatomas

which are supported by clinical and experimental evidence. The diag-nosis of aural cholesteatoma is made on otoscopic examination andspecial imaging procedures, such as high-resolution CT and MRI. Theexpansion of cholesteatomas may result in erosion of surrounding struc-tures, which may result in local and intracranial complications. Cho-

lesteatomas can be eradicated from the temporal bone only by surgicalresection; various surgical approaches are available.The main mechanism of bone resorption in chronic otitis media

with and without cholesteatoma is thought to be the increased numberand activity of osteoclasts. Many metabolites and cytokines are knownto activate osteoclasts locally. PGE2, osteoprotegerin, and macrophage

limited to the anterior petrous apex, and the middle ear is normal.197 Many patients have involvement of cranial nerves V, VI, and VII.

D ag t When the diagnosis of petrous apicitis is suspected on clinical grounds,the most appropriate diagnostic procedure is CT. High-resolution CTusually shows details of the petrous apex and provides important detailsabout potential surgical routes. A pneumatized petrous apex on theuninvolved side can sometimes be contrasted with a uid- lled orsclerotic petrous apex on the involved side, although Roland and col-leagues198 have shown that asymmetry of the petrous apex is not diag-nostic for apicitis because asymmetric pneumatization of the apex canoccur in healthy subjects. If CT scan indicates a potential apicitis, MRImay add information about the nature of the uid or tissue within theapex (Fig. 139-19). A gallium bone scan may provide additional infor-mation, showing increased uptake on the side of the apicitis. A com-bination of MRI and CT is necessary to evaluate normal anatomicvariations and to be capable of differentiating diagnosis.199

Ma ag mThe management of petrous apicitis is directed toward control of theinfection. If topical and systemic antibiotic management is inadequate

ME

C

APAIAC

PPA

Figure 139-18. Pneumatic anterior petrous apex (APA) and posteriorpetrous apex (PPA) in horizontal temporal bone histopathologic section.There is a large air cell in APA and air cells in PPA. C, carotid artery;IAC, internal auditory canal; ME, middle ear. (From Chole RA. Petrousapicitis: surgical anatomy. Ann Otol Rhinol Laryngol . 1985;94:251.)

AA BB

Figure 139-19. CT (A) and MRI (B) are helpful adjuncts in the diagnosis of petrous apicitis. A, In the CT scan, the uid- lled petrous apex (arrow) can be compared with the opposite air- lled apex. B, In the MRI scan, the area of increased signal is present in involved apex (arrow) . An amber-colored, uid- lled cyst was found at surgery.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


16/16


Costerton JW, Stewart PS, Greenberg EP. Bacterial bio lms: a common causeof persistent infections.Science . 1999;284:1318-1322.

Ehrlich GD, Veeh R, Wang X, et al. Mucosal bio lm formation on middle-earmucosa in the chinchilla model of otitis media. JAMA. 2002;287:1710-1715.

Giebink GS, Daly K, Buran DJ, et al. Predictors for postoperative otorrhea following tympanostomy tube insertion. Arch Otolaryngol Head Neck Surg.1992;118:491-494.

Hildmann H, Sudhoff H. Middle Ear Surgery . New York: Springer Verlag;2006.

Hussl B, Lim DJ. Histopathology of tympanosclerosis. In: Lim DJ, BluestoneCD, Klein JO, eds. Recent Advances in Otitis Media with E usion. St. Louis:Mosby; 1984.

Lempert J. Complete apicectomy (mastoidotympano-apicectomy): new tech-nique for complete apical exenteration of apical carotid portion of petrouspyramid. Arch Otolaryngol Head Neck Surg . 1937;25:144.

Lieu JE, Muthappan PG, Uppaluri R. Association of re ux with otitis media inchildren. Otolaryngol Head Neck Surg . 2005;133:357-361.

McKennan KX, Chole RA. Post-traumatic cholesteatoma. Laryngoscope .1989;99(8 Pt 1):779-782.

Michaels L. An epidermoid formation in the developing middle ear: possible

source of cholesteatoma. J Otolaryngol . 1986;15:169-174.Sudhoff H, Linthicum FH Jr. Cholesteatoma behind an intact tympanic mem-

brane: histopathologic evidence for a tympanic membrane origin.Otol Neu-rotol . 2001;22:444-446.

Sudhoff H, Tos M. Pathogenesis of attic cholesteatoma: clinical and immuno-histochemical support for combination of retraction theory and proliferationtheory. Am J Otol . 2000;21:786-792.

colony-stimulating factor are some of the most potent osteoclast-acti-vating metabolites. Cytokines and growth factors such as IL-1, IL-6,tumor necrosis factor- , epidermal growth factor, parathyroid hor-monerelated protein, and TGF- seem to play an important role inmiddle ear infections and cholesteatoma. Tissue-type plasminogen acti-vator and collagenase also may play a role in local invasion of chole-steatoma. More recent evidence suggests that bacterial bio lms withinchronically infected cholesteatomas may contribute to their aggressive-ness by local elaboration of lipopolysaccharide.

Sensorineural hearing loss and tympanosclerosis are thought to becomplications of otitis media. Tympanosclerosis in the middle ear oftenleads to conductive hearing loss caused by ossicular xation. Petrousapicitis may be complicated by the spread of infection within the tem-poral bone into the petrous apex. The appropriate diagnostic procedureis high-resolution CT scanning. MRI and gallium bone scan may provide additional information. The management of petrous apicitis isdirected toward control of the infection and includes topical and sys-temic antibiotic management and surgical approach through many surgical routes.

suGGesteD reADinGs

Brackmann DE, Toh EH. Surgical management of petrous apex cholesterolgranulomas.Otol Neurotol . 2002;23:529-533.

Chole RA. Cellular and subcellular events of bone resorption in human andexperimental cholesteatoma: the role of osteoclasts.Laryngoscope . 1984;94:76-95.

Chole RA, Donald PJ. Petrous apicitis: clinical considerations. Ann Otol Rhinol Laryngol . 1983;92(6 Pt 1):544-551.

Chole RA, Faddis BT. Evidence for microbial bio lms in cholesteatomas. ArchOtolaryngol Head Neck Surg . 2002;128:1129-1133.

For complete list of references log onto www.expertconsult.com.

chronic otitis

Documents