Cholesteatomas are expanding lesions of the temporal bone that are composed of a stratified squamous epithelial outer lining and a desquamated keratin center. The matrix is composed of fully differentiated squamous epithelium resting on a connective tissue matrix. Cholesteatomas were named by Johannes Mueller in 1838 with the original erroneous belief that one of the primary components of the tumor was fat. Cholesteatomeas may develop anywhere within pneumatized portions of the temporal bone, with the most frequent locations being the middle ear and the mastoid. This tumor formation can lead to many sequelae including infection, otorrhea, bone destruction, hearing loss, facial nerve paresis or paralysis, labyrinthine fistula, as well as intracranial complications including epidural and subdural abscesses, parenchymal brain abscesses, meningitis, and thrombophlebitis of the dural venous sinuses. The destructive nature of cholesteatomas can be attributed to the outer layer of granulation tissue that secretes multiple enzymes which come into contact with bone.
Types of Cholesteatoma
Cholesteatomas can be classified as one of two different types: congenital and acquired. Congenital cholesteatomas are believed to arise from embryonal inclusions or rests of epithelial cells. This classification of cholesteatoma presents behind an intact tympanic membrane, without continuity to the external ear canal and in the absence of etiological factors such as tympanic membrane perforation and a history of ear infections. Modification of the definition of congenital cholesteatoma by Levenson, et. al. in 1989, established a set of criteria for the definition of congenital cholesteatoma in the middle ear. These criteria included, a white mass medial to a normal tympanic membrane, a normal pars flaccida and pars tensa, no prior history of otorrhea or perforations, and no prior otologic procedures. In addition, prior bouts of otitis media were not grounds for exclusion as was the case in the original definition. Levenson’s study revealed that the mean age at presentation was 4.5 years with a male preponderance of 3:1. Two thirds of the cases were confined to the anterior superior quadrant of the middle ear. The pathogenesis of congenital cholesteatomas has been theorized by many physicians and researchers, yet the underlying etiology remains unclear. Two prominent theories include the failure of the involution of ectodermal epithelial thickening that is present during fetal development in proximity to the geniculate ganglion and metaplasia of the middle ear mucosa.
Acquired cholesteatomas are subdivided into primary acquired and secondary acquired cholesteatoma. Several pathogenic mechanisms have been proposed to explain the formation of acquired cholesteatomas, with no single process being accepted as the mechanism for the development of all cases. The common factor of all acquired cholesteatomas is that the keratinizing squamous epithelium has grown beyond its normal limits. Primary acquired cholesteatomas ultimately form due to an underlying Eustachian tube dysfunction that causes retraction of the pars flaccida. Eustachian tube dysfunction directly results in poor aeration of the epitympanic space. This draws the pars flaccida medially on top of the malleus neck, forming a retraction pocket. Once a retraction pocket develops, the normal migratory pattern of the tympanic membrane epithelium is altered, enhancing the potential accumulation of keratin. As keratin accumulates, the sac that has formed will slowly enlarge.
The pathogenesis of secondary acquired cholesteatomas is attempted to be explained by several theories: the implantation theory, the metaplasia theory, and the epithelial invasion theory. The implantation theory proposes that squamous epithelium becomes implanted into the middle ear as a result of surgery, a foreign body, or a blast injury. The metaplasia theory proposes that desquamated epithelium is transformed to keratinized stratified squamous epithelium secondary to chronic or recurrent otitis media. This theory is not believed to be an explanation for a significant cause of cholestatoma formation in humans. The mechanism behind the epithelial invasion theory is that whenever there is a permanent perforation of the tympanic membrane, the squamous epithelium starts migrating along the perforation edge and may continue medially along the undersurface of the drum destroying the columnar epithelium. It has been proposed that this process is triggered by lingering, chronic infection within the tympanic cavity. Papillary ingrowth refers to the development of cholesteatoma arising from an intact pars flaccida. It is theorized that an inflammatory reaction in Prussack’s space, likely secondary to poor ventilation in this area, may cause a break in the basal membrane allowing a cord of epithelial cells to start their proliferation inwards.
Cholesteatoma growth patterns are predictable in that they are channeled along characteristic pathways by ligaments, folds, and ossicles. The most common locations from which cholesteatoma arise are the posterior epitympanum, the posterior mesotympanum and the anterior epitympanum. A basic knowledge of the anatomy of this area provides a foundation for understanding the disease progression and concepts for surgical management.
The middle ear can be divided into three compartments: the mesotympanum, hypotympanum, and epitympanum. The boundary that defines these areas is the external auditory canal. The epitympanum is superior and medial to the superior aspect of the external auditory canal. The hypotympanum is inferior and medial to the inferior aspect of the external auditory canal. The mesotympanum is medial to the external auditory canal with its inferior and superior boundaries defined by the inferior and superior aspect of the external auditory canal respectively.
The mesotympanum contains the stapes, long process of the incus, handle of the malleus and the oval and round windows. The eustachian tube exits from the anterior aspect of the mesotympanum. Two recesses extend posteriorly from the mesotympanum that are often not visible directly- the facial recess and sinus tympani. The facial recess and sinus tympani, are the most common location for cholesteatoma persistence after chronic ear surgery. The facial recess is lateral to the facial nerve, bounded by the fossa incudis superiorly and the chorda tympani nerve laterally. This recess may be directly accessed through a posterior approach via the mastoid (posterior tympanotomy or facial recess approach).The sinus tympani lies between the facial nerve and the medial wall of the mesotympanum and is very difficult to access surgically.
The epitympanum lies above the level of the short process of the malleus, containing the head of the malleus, body of the incus, and their associated ligaments and mucosal folds. The annular ligament sends off fibrous bands from the anterior and posterior tympanic spines that meet at the neck of the malleus. The dehiscent area in the tympanic bone, known as the notch of Rivinus, lies above these bands. The dense fibers that form the middle layer of the pars tensa do not extend to the pars flaccida. The lack of this structural support predisposes Shrapnell’s membrane to retraction when negative middle ear pressure is present secondary to Eustachian tube dysfunction.
Cholesteatomas of the epitympanum start in Prussack’s space between the pars flaccida and neck of the malleus with the upper boundary being the lateral mallear fold. The most common spread patterns of cholesteatomas from Prussack’s space are through the posterior epitympanum, posterior mesotympanum and anterior epitympanum. The most common spread pattern of the three is the posterior epitympanic route where the cholesteatoma spreads to the superior incudal space lateral to the body of the incus potentially gaining access to the mastoid through the aditus ad antrum. The second most common is the inferior route, thought the posterior pouch of von Troeltsch. This pouch lies between the tympanic membrane and the posterior mallear fold. Spread via this route allows cholesteatoma to gain access to the regions of the stapes, round window, sinus tympani and facial recess. Anterior epitympanic cholesteatomas form anterior to the malleus head. They may be easily overlooked during tympanomastoidectomy if the area is not explored. Facial nerve dysfunction may occur with these lesions, which can also gain access to the supratubal recess of the middle ear via the anterior pouch of von Troeltsch.
As previously described, the hypotympanum is the portion of the middle ear that lies inferior and medial to the floor of the bony ear canal. It is an irregular bony groove that is seldom involved by cholesteatoma. Occasionally, the jugular bulb may be dehiscent in this area.
Prevention of Cholesteatoma Formation
When a patient evaluated in clinic is noted to have a retraction pocket, the otolaryngologist must recognize that this manifestation is due to Eustachian tube dysfunction and that the condition precedes the development of acquired cholesteatoma. As a result, a long-term tympanostomy tube should be placed to resolve the negative middle ear pressure. This intervention may allow the tympanic membrane to revert to a neutral position. If the retraction pocket is adherent to the ossicles or folds or if it has been present for an extended period of time, the retraction pocket will persist. If the retraction pocket persists, surgical exploration may be indicated.
As always, the initial patient evaluation should include a thorough history. A detailed otologic history should be obtained in order to elicit the early symptoms of cholesteatoma including hearing loss, otorrhea, otalgia, nasal obstruction, tinnitus and vertigo. A previous history of middle ear disease, such as chronic otitis media and/or tympanic membrane perforation may be revealed. Progressive unilateral hearing loss with a chronic foul smelling otorrhea should raise suspicion.
In addition to a thorough head and neck examination, the otologic examination should be meticulous and complete. Otomicroscopy is of the utmost importance in evaluating the presence of cholesteatoma and extent of disease. The ear should be thoroughly cleaned of otorrhea and debris with cotton-tipped applicators or suction. A retraction pocket may be seen, often in the attic and posterosuperior quadrant of the tympanic membrane. Accumulation of squamous debris may occur within the pocket. Granulation tissue may arise from the diseased infected bone of the scutum or posterior bony wall. When extensive, a polyp may protrude through an attic defect. Extreme caution should be used with polyp removal as it may be adherent to important underlying structures such as the ossicles or facial nerve. Pneumatic otoscopy should be performed in every patient with a cholesteatoma. A positive fistula (pneumatic otoscopy will result in nystagmus and vertigo) response suggests erosion of the semicircular canals or cochlea. Cultures should be obtained with wet, infected ears. Topical and/or oral antibiotics should be administered in these cases.
Pure tone audiometry with air and bone conduction, speech reception thresholds, and word recognition usually reveal a conductive hearing loss in the affected ear. The degree of conductive loss will vary considerably depending on the extent of disease. A moderate conductive deficit in excess of 40 dB indicates ossicular discontinuity, usually from erosion of the long process of the incus or capitulum of the stapes. A mild conductive deafness may be present with extensive disease if the cholesteatoma sac transmits sound directly to the stapes or footplate. Audiometry results should always be correlated with the 512Hz tuning fork exam. Tympanometry results will vary and may suggest decreased compliance or perforation of the tympanic membrane.
Preoperative imaging with computed tomographies (CTs) of the temporal bones (2mm -section without contrast in axial and coronal planes) allows for evaluation of anatomy, which may reveal evidence of the extent of the disease as well as screen for asymptomatic complications. Although a temporal bone CT is not essential for preoperative evaluation, they should be obtained for revision cases due to altered landmarks from previous surgery, for patients with complications of chronic suppurative otitis media, suspected congenital abnormalities, or cases of cholesteatoma in which sensorineural hearing loss, vestibular symptoms, or other evidence of complications exist.
Preoperative counseling is an absolute necessity prior to surgery. The primary objective of surgery is a safe dry ear which is accomplished by treating all supervening complications, removing diseased bone, mucosa, granulation polyps, and cholesteatoma while preserving as much normal anatomy as possible. Improvement of hearing is a secondary goal. Possible adverse outcomes must be discussed including facial paralysis, vertigo, further hearing loss, and tinnitus. The patient should understand that long-term follow-up will be necessary and that they may need additional surgeries.
Cholesteatoma is treated surgically with a primary goal of total eradication of cholesteatoma to obtain a safe and dry ear. The second objective is restoration or maintaining the functional capacity of hearing. The third objective is to maintain a normal anatomic appearance of the ear if possible. The surgical procedure to be used should be designed for each individual case according to the extent of disease. More extensive disease will usually dictate a more aggressive surgical approach.
Canal-Wall-Down (CWD) Procedure
Prior to the advent of the tympanoplasty, all cholesteatoma surgery was performed using a CWD approach. This procedure involves taking down the posterior canal wall to the level of the vertical facial nerve and exteriorizing the mastoid into the external ear canal. The epitympanum is obliterated with removal of the scutum, head of the malleus and incus. A classic CWD operation is the modified radical mastoidectomy in which the middle ear space is preserved. The radical mastoidectomy is a CWD operation in which the middle ear space is eliminated and the eustachian tube is plugged. Meatoplasty should be large enough to allow good aeration of the mastoid cavity and permit easy visualization to facilitate postoperative care and self cleaning. The indications for this as an initial approach are cholesteatoma in an only hearing ear, significant erosion of the posterior bony canal wall, history of vertigo suggesting a labyrinthine fistula, recurrent cholesteatoma after canal-wall-up surgery, poor eustachian tube function, and a sclerotic mastoid with limited access to the epitympanum.
The advantages of the CWD procedure are that residual disease is easily detected, recurrent disease is rare, and the facial recess is exteriorized. The major disadvantage of this procedure is the open cavity and that mastoid bowl maintenance can be a lifelong problem. Healing takes longer in open cavities and the middle ear is shallow and difficult to reconstruct. Dry ear precautions are essential.
Canal-Wall-Up (CWU) Procedure
The CWU procedure was developed to avoid the problems and maintenance necessary when CWD procedures are performed. CWU consists of preservation of the posterior bony external auditory canal wall during simple mastoidectomy with or without a posterior tympanotomy. A staged procedure is often necessary with a scheduled second look operation at 6 to 18 months for removal of residual cholesteatoma and ossicular chain reconstruction if necessary. The procedure should be adapted to the extent of disease as well as the skill of the otologist. This approach may be indicated in patients with a large pneumatized mastoid and a well aerated middle ear space, suggesting good eustachian tube function. CWU procedures are contraindicated in only hearing ears or in patients with a labyrinthine fistula, long-standing ear disease, or poor eustachian tube function.
The advantages of CWU compared with CWD mastoidectomies are more rapid healing time, easier long-term care, no dry ear precautions, and hearing aids are easier to fit and wear if they are needed. The disadvantages associated with this procedure are the difficulty of technique leading to longer operative time, residual disease is more difficult to detect, retraction pockets leading to recurrent disease are possible, and staged operations are often necessary.
Transcanal Anterior Atticotomy
A transcanal anterior atticotomy is indicated for limited cholesteatoma involving the middle ear, ossicular chain, and epitympanum. If the extent of the cholesteatoma is unknown, this approach can be combined with a CWU mastoidectomy or extended to a CWD procedure. The atticotomy involves elevation of a tympanomeatal flap via an endaural incision with removal of the scutum to the limits of the cholesteatoma. After removal of the disease, the aditus is obliterated with muscle, fascia, cartilage or bone prior to reconstruction of the middle ear space. Some advocate reconstruction of the lateral attic wall with bone or cartilage, however, this may lead to retraction disease and possible recurrence in patients with poor eustachian tube function.
Bondy Modified Radical Mastoidectomy
Although rarely used today, the Bondy procedure is a useful for specific types of cholesteatoma. It is indicated for attic and mastoid cholesteatoma that does not involve the middle ear space and is lateral to the ossicles. Preferably, the mastoid should be poorly developed for creation of a small cavity. The eustachian tube function should be adequate, with an intact pars tensa and aerated middle ear space. The Bondy procedure is performed like the modern modified radical mastoidectomy with the exception that the middle ear space is not entered.
The expansion of cholesteatomas combined with the propensity of infection result in numerous complications that include ossicular chain destruction, exposure of the membranous labyrinth, exposure of the facial nerve and dura, and infection of the mastoid and intracranial spaces.
Conductive hearing loss is a common complication of cholesteatoma as ossicular chain erosion occurs in 30% of cases. Erosion of the lenticular process and or stapes superstructure may produce a conductive hearing loss as high as 50dB. However, hearing loss may vary with the development of a natural myringostapediopexy or transmission of sound through a cholesteatoma sac to the stapes or footplate. This results in less of a conductive hearing loss. The ossicular chain should always be assumed to be intact. Evidence of sensorineural hearing loss may indicate involvement of the labyrinth. Following surgery, 3% of operated ears have further impairment permanently due to the extent of the disease present or due to complications in the healing process. Patients should be counseled that there is a possibility of total loss of hearing in the operated ear. Also, with two-staged operations, the hearing will be worse after the first operation.
A labyrinthine fistula may occur in as many as 10% of patients with chronic ear infection due to cholesteatoma. A fistula should be suspected in a patient with longstanding disease with sensorineural hearing loss and/or vertigo induced by noise or pressure changes in the middle ear. Absence of a positive fistula test does not rule out this complication. Fine cut CT of the temporal bone should be obtained if this condition is suspected. The most common site of a labyrinthine fistula is the horizontal semicircular canal, although the basal turn of the cochlea is also at risk. The procedure of choice with this complication is a modified radical mastoidectomy, as discussed previously. Management of the matrix overlying the fistula depends on the infection status of the ear, the degree of hearing loss in the affected and nonaffected ear, the size and location of the fistula, and the surgeon’s experience. In an only hearing ear, matrix should be left intact over the fistula. Matrix should also be left over extensive fistulae of the vestibule or cochlea if hearing is normal. Matrix can be removed in a relatively dry, uninfected ear with a normal hearing opposite ear, and the fistula covered with bone pate or fascia.
Facial paralysis in patients with cholesteatoma requires immediate surgery. The paralysis may develop acutely secondary to infection or slowly from chronic expansion of the cholesteatoma. A temporal bone CT should be obtained to help localize the nerve involvement. The most common site of facial nerve involvement is at the geniculate ganglion due to disease in the anterior epitympanum. A simple mastoidectomy with a facial recess approach will expose the tympanic and mastoid portions of the facial nerve, while a middle fossa approach is required when there is petrous apex involvement. Removal of the cholesteatoma and infected material with decompression of the nerve usually results in the recovery of facial nerve function. Administration of intravenous antibiotics and high-dose steroids are also helpful. Iatrogenic injury to the nerve during surgery should be immediately repaired with decompression of the nerve proximal and distal to the site of injury.
Intracranial complications of cholesteatoma are potentially life-threatening. Infections such as a periosteal abscess, lateral sinus thrombosis and intracranial abscess occur in less than 1% of all cholesteatomas. Findings suggesting an impending intracranial complication include suppurative malodorous otorrhea, usually with chronic headache, pain and/or fever. The presence of mental status changes with nuchal rigidity or cranial neuropathies warrant neurosurgical consultation with urgent intervention. Epidural abscess, subdural empyema, meningitis and cerebral abscesses should be treated immediately prior to definitive otologic management of ear disease.
A brain hernia presents at a revision surgery as a meningoencephalocele or an encephalocele. It results from a defect in the tegmen tympani or tegmen mastoideum due to trauma from the drill at the time of the original surgery. This condition must be carefully inspected and repaired at the time of surgery.
The pathogenesis of cholesteatoma remains uncertain. The identification and behavior of the disease, however, is well described. For successful management of cholesteatomas, it is essential to possess a basic knowledge of the important anatomic and functional characteristics of the middle ear. Careful and thorough evaluations are the key to the early diagnosis and treatment of the disease. Early diagnosis and treatment can prevent complications and preserve hearing. Treatment of cholesteatoma is surgical with the primary goal to eradicate disease and provide a safe and dry ear. Surgical approaches must be customized to each patient depending on the extent of their disease. The surgeon must be aware of the serious and potentially life-threatening complications of cholesteatomas.
Cholesteatoma is an abnormal accumulation of keratin-producing squamous epithelium in the middle ear, epitympanum, mastoid or petrous apex. It has been further defined as a three dimensional epidermoid structure exhibiting independent growth, replacing middle ear mucosa, and resorbing underlying bone. Although it is not a neoplastic lesion, it can be insidious and potentially dangerous to the patient. The term “cholesteatoma” was first used by Johannes Müller in 1838 to describe a true neoplasm he thought was “a pearly tumor of fat…among sheets of polyhedral cells”. In fact, the cholesteatoma appears histologically as a benign keratinizing squamous cell cyst made up of three components, i.e. the cystic content, the matrix and the perimatrix. The cystic content is composed of fully-differentiated anucleate keratin squames. The matrix contains the keratinizing squamous epithelium lining a cyst-like structure. The perimatrix or lamina propria is the peripheral part of the cholesteatoma consisting of granulation tissue, which may contain cholesterol crystals. The perimatrix layer is in contact with bone, and it is this granulation tissue, which produces various proteolytic enzymes that may result in bone destruction.
II. Classification and pathogenesis
Cholesteatoma may be classified according to presumed etiology into two general categories: congenital and acquired. Acquired cholesteatomas can be further divided into primary and secondary acquired. Congenital cholesteatomas are thought to arise from embryonal inclusions or rests of epithelial cells. It refers to cholesteatomas present behind an intact tympanic membrane, without continuity to the external ear canal and in the absence of etiological factors such as tympanic membrane perforation and a history of ear infections. They can be further classified according to location within the temporal bone (the petrous pyramid, mastoid and middle ear cleft). Levenson, et. al., established a set of criteria for the definition of congenital cholesteatoma in the middle ear. These included, a white mass medial to a normal tympanic membrane, a normal pars flaccida and pars tensa, no prior history of otorrhea or perforations, and no prior otologic procedures. In addition, prior bouts of otitis media were not grounds for exclusion. In their study (over 40 cases), the mean age at presentation was 4.5 years with a male preponderance of 3:1. Two thirds of the cases were confined to the anterior superior quadrant of the middle ear.
Several pathogenic mechanisms have been produced to explain the development of acquired cholesteatomas. No single process is accepted as the mechanism for the development of all cases. However, in all types the keratinizing squamous epithelium has spread beyond its normal limits. With primary acquired cholesteatomas, the cause is due to underlying Eustachian tube dysfunction resulting in retraction of the pars flaccida. The problem becomes poor aeration of the epitympanic space which draws the pars flaccida medially on top of the malleus neck. Once a retraction pocket develops, the normal migratory pattern of the tympanic membrane epithelium is altered, encouraging the accumulation of keratin. If not addressed, the sac slowly enlarges to and around the ossicles, the attic walls, etc. The following theories explain secondary acquired cholesteatoma pathogenesis. The implantation theory proposes that squamous epithelium becomes implanted into the middle ear as a result of surgery, foreign body (ventilating tubes), or blast injury. The metaplasia theory explains that as a result of chronic or recurrent otitis media the low cuboidal epithelium of the middle ear becomes transformed to a keratinized stratified squamous epithelium, similar to other parts of the body (nose, sinuses, bronchi) in response to chronic irritation or infection. The mechanism behind the epithelial invasion or migration theory is that whenever there is a permanent perforation of the tympanic membrane, the squamous epithelium starts migrating along the perforation edge and may continue medially along the undersurface of the drum destroying the columnar epithelium. It has been proposed that this process is triggered by lingering, chronic infection within the tympanic cavity. Papillary ingrowth refers to the development of cholesteatoma arising from an intact pars flaccida (Shrapnell’s membrane). It is theorized that an inflammatory reaction in Prussack’s space, likely secondary to poor ventilation in this area, may cause a break in the basal membrane allowing a cord of epithelial cells to start their proliferation inwards.
III. Anatomic Considerations
Cholesteatomas enlarge with fairly typical patterns of growth. The most common locations from which cholesteatoma arise are the posterior epitympanum, the posterior mesotympanum and the anterior epitympanum. Cholesteatomas are channeled along characteristic pathways by surrounding mucosal folds, the middle ear ossicles, and their suspensory ligaments. A basic knowledge of the anatomy of this area provides a foundation for understanding the disease progression and concepts for surgical management.
The middle ear can be divided into three compartments: the mesotympanum, hypotympanum, and epitympanum. The mesotympanum contains the stapes, long process of the incus, handle of the malleus and the oval and round windows. The eustachian tube exits from the anterior aspect of the mesotympanum. Two recesses extend posteriorly from the mesotympanum that are often impossible to visualize directly. These spaces, the facial recess and sinus tympani, are the most common location for cholesteatoma persistence after chronic ear surgery. The sinus tympani lies between the facial nerve and the medial wall of the mesotympanum and is very difficult to access surgically. The facial recess is lateral to the facial nerve, bounded by the fossa incudis superiorly and the chorda tympani nerve laterally. It may be directly accessed via a posterior approach, through the mastoid (posterior tympanotomy or facial recess approach). The hypotympanum is the portion of the middle ear that lies below the floor of the bony ear canal. It is an irregular bony groove that is seldom involved by cholesteatoma. Occasionally, the jugular bulb may be dehiscent in this area.
The epitympanum lies above the level of the short process of the malleus, containing the head of the malleus, body of the incus, and their associated ligaments and mucosal folds. The annular ligament sends off fibrous bands from the anterior and posterior tympanic spines that meet at the neck of the malleus. The dehiscent area in the tympanic bone, known as the notch of Rivinus lies above these bands. The dense fibers that form the middle layer of the pars tensa do not extend to the pars flaccida. The lack of this structural support predisposes Shrapnell’s membrane to retraction in the face of negative middle ear pressure.
Epitympanic cholesteatomas start in Prussack’s space between the pars flaccida and neck of the malleus with the upper boundary of the lateral mallear fold. The most common locations of spread of cholesteatomas from Prussack’s space are via the posterior epitympanum, posterior mesotympanum and anterior epitympanum, in that order. The most common is the posterior epitympanic route where the cholesteatoma spreads to the superior incudal space lateral to the body of the incus potentially gaining access to the mastoid through the aditus ad antrum. The second most common is the inferior route, thought the posterior pouch of von Troeltsch. This pouch lies between the tympanic membrane and the posterior mallear fold. Spread via this route allows cholesteatoma to gain access to the regions of the stapes, round window, sinus tympani and facial recess. Anterior epitympanic cholesteatomas form anterior to the malleus head. They may be easily overlooked during tympanomastoidectomy if not explored. Facial nerve dysfunction may occur with these lesions, which can also gain access to the supratubal recess of the middle ear via the anterior pouch of von Troeltsch.
History – A careful otologic history should be obtained in order to elicit the early symptoms of cholesteatoma. The most common presenting symptoms are hearing loss, otorrhea, otalgia, nasal obstruction, tinnitus and vertigo. A previous history of middle ear disease, such as chronic otitis media and/or tympanic membrane perforation may be evident. Progressive unilateral hearing loss with a chronic foul smelling otorrhea should raise suspicion.
Physcial Examination - In addition to a through head and neck examination, particular attention should be pain to the otologic exam. Otomicroscopy is most important in evaluating the presence of cholesteatoma and extent of disease. The ear should be thoroughly cleaned of otorrhea and debris. A retraction pocket may be seen, often in the attic or posterosuperior quadrant of the TM. Accumulation of squamous debris may occur within the pocket. Granulation tissue may be arise from the diseased infected bone of the scutum or posterior bony wall. When extensive, a polyp may protrude through an attic defect. Extreme caution should be used with polyp removal as it may be adherent to important underlying structures such as the ossicles or facial nerve. Pneumatic otoscopy should be performed in every patient with a cholesteatoma. A positive fistula response suggests erosion of the semicircular canals or cochlea. Cultures should be obtained with wet, infected ears. Topical and/or oral antibiotics should be administered in these cases.
Audiology: Pure tone audiometry with air and bone conduction, speech reception thresholds, and word recognition usually reveal a conductive hearing loss in the affected ear. The degree of conductive loss will vary considerably depending on the extent of disease. A moderate conductive deficit in excess of 40 dB indicates ossicular discontinutiy, usually from erosion of the long process of the incus or capitulum of the stapes. A mild conductive deafness may be present with extensive disease if the cholesteatoma sac transmits sound directly to the stapes or footplate. Audiometry resluts should always be correlated with the 512Hz tuning fork exam. Tympanometry resulsts will vary and may suggest decreased compliance or perforation of the tympanic membrane.
Imaging – Preoperative imaging with CT of the temporal bones allows pre-operative imaging of anatomy, some evidence of the extent of the disease and a screen for asymptomatic complications. However. it has not gained wide acceptance as an essential aid to planning surgery in uncomplicated cases of the cholesteatoma. Temporal bone CT should be obtained for revision cases due to altered landmarks from previous surgery. for patients with compliccations of chronic suppurative otitis media, suspected congenital abnormalities, or cases of cholesteatoma in which sensorineural hearing losss vestibular symptoms or other evidence of complications exist.
Cholesteatoma is a surgical disease for which the primary, universally accepted goal is total eradication of cholesteatoma to obtain a safe, dry ear. The second objective is restoration or maintaining the functional capacity of the ear, the hearing. The third objective is to maintain a normal anatomic appearance of the ear if possible. Management of complications when they arise takes priority over other objectives. The surgical procedure to be used should be designed for each individual case according to the pathology present. The extent of disease often will determine the aggressiveness of the surgical approach.
As with any surgical procedure, preoperative counseling is mandatory. Surgical goals, risks of surgery (facial paralysis, vertigo, tinnitus, hearing loss), possibility of staged procedure, need for long-term follow-up and routine aural toilet if necessary should be reviewed in detail with the patient.
Medical management, including aggressive aural toilet, powder applications, and office local care may exteriorize and safely decompress the accumulating keratin debris. This may be a valid management strategy for patients in whom anesthesia poses an unacceptable risk. Such management is not recommended in children. Preoperatively, it is very important to eliminate drainage and any acute inflammatory changes. This will reduce troublesome intraoperative bleeding and help with the delineation of irreversible disease that must be removed from preservable structures.
Canal-wall-down (CWD) procedures
Prior to the advent of tympanoplasty techniques, all cholesteatoma surgery was of this type. These procedures involve taking down the posterior canal wall to the vertical facial nerve, exteriorizing the mastoid into the external ear canal. The epitympanum is obliterated with removal of the scutum, head of the malleus and incus. A classic CWD operation is the modified radical mastoidectomy in which the middle ear space is preserved. The radical mastoidectomy is a CWD operation in which the middle ear space is eliminated and the eustachian tube plugged. Meatoplasty should be large enough to allow good aeration of the mastoid cavity and permit easy visualization to facilitate postoperative care and self cleaning. The indications for this as an initial approach are:
- cholesteatoma in an only hearing ear
- significant erosion of the posterior bony canal wall
- history of vertigo suggesting a labyrinthine fistula
- recurrent cholesteatoma after ICW surgery with poor eustachian tube function
- sclerotic mastoid (with limited access to the epitympanum)
The advantages of the CWD procedure are that residual disease is easily detected, recurrent disease is rare, and the facial recess is exteriorized. The major disadvantage of this procedure is the open cavity and that mastoid bowl maintenance can be a lifelong problem. Healing takes longer in open cavities and the middle ear is shallow and difficult to reconstruct. Also, dry ear precautions are necessary.
Intact-canal-wall (ICW) procedure
This procedure was developed to avoid cavity problems altogether. It consists of preservation of the posterior bony external auditory canal wall during simple mastoidectomy with or without a posterior tympanotomy. A staged procedure is often necessary with a scheduled second look operation at 6 to 12 months for removal of residual cholesteatoma and ossicular chain reconstruction. The procedure should be adapted to the extent of disease as well as the skill of the otologist. This approach may be indicated in patients with a large pneumatized mastoid and a well aerated middle ear space, suggesting good eustachian tube function. Intact canal wall procedures are contraindicated in only hearing ears or in the patient with a labyrinthine fistula, long-standing ear disease, or poor eustachian tube function.
The advantages of this procedure compared with CWD mastoidectomies are more rapid healing time, easier long-term care, no water precautions necessary and hearing aids should they be needed are easier to fit and wear. The disadvantages associated with this procedure include the difficulty of technique with more operative time generally, residual disease is more difficult to detect, retraction pockets leading to recurrent disease are possible, and staged operations are often necessary.
Transcanal anterior atticotomy
This procedure is indicated for limited cholesteatoma involving the middle ear, ossicular chain, and epitympanum. If the extent of the cholesteatoma is unknown, this approach can be combined with an intact canal wall mastoidectomy or extended to a CWD procedure. The atticotomy involves elevation of a tympanomeatal flap via an endaural incision with removal of the scutum to the limits of the cholesteatoma. After removal of the disease, the aditus is obliterated with muscle, fascia, cartilage or bone prior to reconstruction of the middle ear space. Some advocate reconstruction of the lateral attic wall with bone or cartilage, however, this may lead to retraction disease and possible recurrence in patients with poor eustachian tube function.
Bondy modified radical mastoidectomy
Although rarely used today, this is a useful procedure for specific types of cholesteatoma. It is indicated for attic and mastoid cholesteatoma that does not involve the middle ear space and is lateral to the ossicles. Preferably, the mastoid should be poorly developed for creation of a small cavity. The eustachian tube function should be adequate, with an intact pars tensa and aerated middle ear space. The Bondy procedure is performed like the modern modified radical mastoidectomy with the exception that the middle ear space is not entered..
VI. Complications of cholesteatoma
Conductive hearing loss is a common complication of cholesteatoma as ossicular chain erosion occurs in as many as 30% of cases. Erosion of the lenticular process and or stapes superstructure may produce a conductive hearing loss as high as 50dB. However, hearing loss may vary with the development of myringostapediopexy or transmission of sound through a cholesteatoma sac to the stapes or footplate. The ossicular chain should always be assumed to be intact. Evidence of sensorineural hearing loss may indicate involvement of the labyrinth. Following surgery, 3% of operated ears have further impairment permanently due to the extent of the disease present or due to complications in the healing process. Patient’s should be counseled that on occasion there is a total loss of hearing in the operated ear, and with two-staged operations, the hearing will be worse after the 1st operation.
Labyrinthine fistula may occur in as many as 10% of patients with chronic ear infection due to cholesteatoma. A fistula should be suspected in a patient with longstanding disease with sensorineural hearing loss and/or vertigo induced by noise or pressure changes in the middle ear. Absence of a positive fistula test does not rule out this complication. Fine cut CT of the temporal bone should be obtained. The most common site is the horizontal semicircular canal, although the basal turn of the cochlea is also at risk. The procedure of choice with this complication has been the modified radical mastoidectomy, as discussed previously. Management of the matrix overlying the fistula depends on the infection status of the ear, degree of hearing loss in the affected and nonaffected ear, size and location of the fistula and surgeon’s experience. In an only hearing ear, matrix should be left intact over the fistula. Matrix should also be left over extensive fistulae of the vestibule or cochlea if hearing is normal. Matrix can be removed in a relatively dry, uninfected ear with a normal hearing opposite ear, and the fistula covered with bone pate or fascia.
Facial paralysis in patients with cholesteatoma requires immediate surgery. The paralysis may develop acutely following infection or slowly from chronic expansion of the cholesteatoma. A CT of the temporal bone is obtained which helps localize the involvement. The most common site is the geniculate ganglion from disease in the anterior epitympanum.
A simple mastoidectomy with facial recess approach will expose the tympanic and mastoid portions of the facial nerve, while a middle fossa approach is required with involvement of the petrous apex. Removal of cholesteatoma and infected material with decompression of the nerve usually suffice. Administration of intravenous antibiotics and high-dose steroids are also helpful. Iatrogenic injury to the nerve during surgery should be immediately repaired with decompression of the nerve proximal and distal to the site of injury.
Intracranial complications of cholesteatoma are potentially life-threatening. Infections such as periosteal abscess, lateral sinus thrombosis and intracranial abscess occur in less than 1% of all cholesteatomas. Findings suggesting an impending intracranial complication include suppurative malodorous otorrhea, usually chronic with headache, pain and/or fever. The presence of mental status changes with nuchal rigidity or cranial neuropathies warrant neurosurgical consultation with urgent intervention. Epidural abscess, subdural empyema, meningitis and cerebral abscesses should be treated immediately prior to definitive otologic management of ear disease.
The exact mechanism or pathogenesis of cholesteatoma is not clearly identified, however, neither the aggressiveness of the disease nor the description of it’s key elements are debated. For successful management of the disease, it is essential to possess a basic knowledge of the important anatomic and functional characteristics of the middle ear. Careful and thorough evaluation are keys to the early diagnosis and treatment of the disease, which may prevent complications and preserve hearing. Cholesteatoma is a surgical disease with the primary goal to eradicate disease and provide a safe, dry ear. The surgical strategies, however, vary greatly depending on the extent of disease and surgeon’s experience. The surgeon must be aware of the serious and potentially life-threatening complications of cholesteatomas.
History of Cochlear Implants
Volta, in the year 1790, became the first person to experience and publish the effects of electrical current on the auditory system. He inserted a metal rod in each ear and then subjected himself to approximately 50 volts of electricity. He reported that the sensation was that of receiving a blow to the head followed by the sound of thick soup boiling. Volta was followed by a string of scientists who continued to experiment with electricity and hearing over the next 167 years. It was Djourno and Eyries who reported the first stimulation of the acoustic nerve by direct application of an electrode in a deaf person (1957). The patient was undergoing an operation for cholesteatoma and the auditory nerve was exposed. An electrode was placed on the nerve and an induction coil and ground electrode were placed in the temporalis muscle. The coil could then be stimulated by currents produced by a second coil placed against the overlying skin. On subsequent experimentation with the patient reported hearing sounds like crickets or a roulette wheel when the second coil was applied. He was able to distinguish simple words and noted improvement of his speech reading ability. This initial implant was followed by a string of implantations performed by House, Doyle, Simmons, and others. Advances in microelectronics, biocompatible materials, and microscopic otologic surgery propelled House to produce the first single-channel implant in 1972 which stimulated the auditory nerve via the scala tympani. In 1984 the cochlear implant gained FDA approval for use in adults. This corresponded with the introduction of multichannel implants which significantly improved spectral perception and open-set speech understanding. The 1990’s saw significant improvements in speech processor designs. SPEAK and CIS speech processing strategies produced large improvements in recipient’s speech recognition. The late 1990’s and early 2000’s saw technologic improvements such as the peri-modiolar contour electrode, split electrodes, behind-the-ear processors, and implantation for children as young as 12 months.
Components of a Cochlear Implant
A cochlear implant consists of several components. All implants have microphones, external speech processors, signal-transfer hardware, transmitters, receivers, and electrodes. Each plays an important part in converting sound to an electrical stimulus. The microphone simply receives and transduces sound into an electrical representation. This is done in an analog (continuous) fashion.
The external speech processor and signal-transfer hardware shapes the electrical signal. This requires amplifying, compressing, filtering, and shaping. Amplification is necessary to increase some signal levels to the point that they can be used in the electrical circuits. Compression is a necessary second step of signal modulation. The normal human ear can hear gradations of sound intensity in a range of 120 dB. Persons with severe to profound hearing loss do not have this same range. In the high frequencies their dynamic range (the difference between their absolute threshold and painful sound) can be only 5 dB! The range in the lower frequencies is often 10-25dB. This means that significant compression of the sound energy must take place in order to render it useful. Thus, all cochlear implants employ gain control of one kind or another. These systems monitor the output voltage and adjust the ratio of compression to keep the output in a range where it provides useful, but not painful stimuli.
Filtering of the input signal is the next step. Frequencies between 100 Hz and 4000 Hz are generally those most important for understanding speech. Sound energy is analyzed using several different types of filters. This allows the unimportant frequencies to be removed and the frequencies of interest to be separately modified. Useful sound information is filtered into frequency bands. This information can then be analyzed for speech patterns and channeled to the appropriate portion of the electrode array.
The transmitter, or outer coil, is placed on the mastoid (usually held in place by magnets) and sends the processed signal to the receiver via radiofrequency. The receiver, surgically placed in a well over the mastoid, receives the signal and sends electrical energy to one or many electrodes in the array. The electrode array, which lies within the cochlea, delivers the electric signal to electrodes along its length. The electrical field generated at these locations serves to discharge the neural components of the auditory system. The eighth nerve then conveys this stimulus to the central nervous system which decodes and interprets the signal.
Just as important as any of the man-made components is the individual’s ability to adjust to, interpret and respond to the electrical stimulus. Length of time spent without sound stimulation of the auditory system, presence or absence of previous experience with sound, personal motivation, community or family support, and opportunities for rehabilitation have been shown to be important factors in achieving a good outcome. These factors likely are important in understanding significant differences in patient outcomes despite similar preoperative auditory deficits, surgical course, and cochlear implant hardware.
Types of Cochlear Implants
Cochlear implants differ in the way that they process sound and how they present electricity to the hearing nerve. Other than the speech processing strategies discussed below, there are two different ways of encoding sound information. The first form, analog coding, involves continuous coding of the sound signal with subsequent transfer to the receiver in multiple radio-frequency channels. Electrodes are continuously stimulated. The second form, digital coding, requires sampling of the sound waveform and assigning a number to these “bits” of information. These bits of information are then transferred to the receiver where they are decoded. Electrodes are stimulated in a pulse fashion. Interestingly, neither approach is 100% effective for all implant users. Recently, combining the two schemes has seen some success.
Cochlear implants can also be distinguished by their use of single vs. multiple channels, the number of electrodes, and their use of either monopolar or bipolar stimulation. The number of electrodes stimulated with different electrical stimuli determines the “channels” used. In other words, an implant may have multiple electrodes, but if the same information is presented to all the electrodes at one time they are essentially functioning as a single channel system. In contrast, multi-channel devices provide different information to several electrodes or groups of electrodes. Early implants had only one electrode (and one channel); recent advances have lead to the development of implants with multiple electrodes (22) and multiple channels (usually 4-8). Having more electrodes means that multiple channels can be localized to areas of the cochlea that are most responsive, and stray current that is stimulating adjacent structures (facial nerve, vestibular nerve) can be rerouted.
Cochlear implants can employ monopolar or bipolar stimulation. In a monopolar system there is only one ground electrode for all the others. The ground is usually located at or outside the round window. Thus an electrical field is created from the stimulated electrode to the ground. A bipolar arrangement is such that the ground for each electrode is much closer (adjacent to, or a few electrodes away). In the highly conductive environment of the inner ear, monopolar stimulation results in some limitations. As additional electrodes are stimulated with different streams (channels) of information the electrical fields created by stimulated electrodes may interfere with fields at other sites. This makes it difficult to stimulate more than one electrode at a time, or electrodes that are close together. The bipolar configuration was an attempt to limit this interaction by placing a ground near each electrode such that a smaller field would be created with less interference and more discrete stimulation. Once again, one approach does not achieve satisfaction with all patients. As a result, many implants offer both grounding methods.
Speech Processing Strategies
There are many different ways of processing the auditory signal for presentation at the level of the cochlear ganglia. The most commonly employed are the spectral peak (SPEAK), continuous interleaved sampling (CIS), and compressed analog stimulation. The SPEAK strategy is characterized by filtering sound into 20 different bands covering the range of 200 Hz to 10,000 Hz. Each filter corresponds to an electrode on the array. The outputs for each filter are analyzed and those channels of highest amplitude that contain speech frequencies are stimulated. The stimulus rate is equal to the period of the lowest frequency of speech (F0). The dominant speech frequency between 280 and 1000 Hz (F1) is then identified and the appropriate apical electrode is stimulated. The dominant speech frequency between 800 and 4000 Hz (F2) is then identified and the appropriate basal electrode is stimulated. Three additional high frequency filters measure input in the 2000-2800 Hz, 2800-4000 Hz, and >4000 Hz ranges. Stimulus is sent to apical electrodes (in order to take advantage of the greater incidence of ganglion cell survival at the apex of the cochlea). These channels provide additional cues for consonant perception and environmental sounds. Electrodes are stimulated sequentially, and at amplitudes specific for each frequency peak.
The continuous interleaved sampled (CIS) strategy is employed by the Clarion and MED-EL systems. This system works by filtering the speech into eight bands. The bands with the highest amplitude within the speech frequencies are subsequently compressed and their corresponding electrodes are stimulated. The CIS strategy uses high-rate pulsatile stimuli to capture the fine temporal details of speech.
The advanced combined encoder (ACE) strategy filters speech into a set number of channels and then selects the highest envelope signals for each cycle of stimulation. Stimulation is carried out in a very rapid fashion (much faster than the SPEAK strategy which stimulates at the rate of the lowest frequency of speech—180-300 cycles/second).
The simultaneous analog strategy (SAS) closely mimics the normal ear. All incoming sound is compressed and filtered into eight channels. These channels are then simultaneously and continuously presented to the appropriate tonotopic electrode. There is no effort to select for speech frequencies. Intensity is coded by either stimulus amplitude, rate or both.
The SAS strategy has met with limited success, whereas the SPEAK and CIS strategies have been relatively successful. It appears that no one system is effective for all recipients. For this reason, recent advances have made it possible for one cochlear implant to offer several speech processing strategies in the same implant. This allows the audiologist and patient to choose what strategy is best for that individual. Currently, the Nucleus systems are made to employ several processing strategies. These include spectral peak (SPEAK), advanced combined encoder (ACE), and continuous interleaved sampling (CIS). The Clarion systems use CIS to stimulate in a monopolar fashion as well as simultaneous analog stimulation (SAS). Medical Electronic (Med-El) produces a product (currently in USA clinical trials) with 12 electrode pairs suitable for deep insertion that relies on the CIS strategy with the most rapid stimulation rate of all implants. Recent advances in technologies have included the development of curved electrode arrays which are intended to more closely approximate the modiolas. Studies seem to indicate that electrodes closer to the basilar membrane need less current to stimulate the nerve and may improve spatial specificity of stimulation.
Indications for Cochlear Implantation
Cochlear implants are FDA approved for adults 18 years and older (no upper age limit) and children age 12 months to 17 years 11 months. Initially implants were only approved for adults who were postlingually deaf and had no improvement with high-powered hearing aids. This group of people has consistently been shown to be benefited by implantation. As more was learned about the benefits of cochlear implantation, the criteria were relaxed. Now, adult criteria include bilateral severe-to-profound sensorineural hearing loss with 70 dB pure tone average, little or no benefit from hearing aids (must attempt binaural high-powered hearing aids for at least 6 months), and psychological suitability. Audiologic examination should show word discrimination scores less than 40% in the best aided condition. The patient should have no anatomical deformity that would preclude implantation success. Finally, the patient should have no physical condition that would preclude a general anesthetic.
Pediatric implantation is indicated in children 12 months or older with bilateral severe-to-profound sensorineural hearing loss with pure tone averages of 90 dB or greater in the better ear. The child must have had no appreciable benefit with hearing aids (evaluated with parental survey when younger than 5 and 30% or less on sentence recognition tests under best-aided conditions when 5 years old or older). Children must tolerate wearing hearing aids for a period (as all cochlear implants have external components), and show some aided communication ability. Children must be enrolled in educational programs that support aural/oral learning and have no medical contraindications. Parents must be highly motivated and have reasonable expectations.
Contraindications for Cochlear Implantation
Not all patients with sensorineural hearing loss are good candidates for cochlear implantation. For example, patients with pure tone thresholds greater than 90 dB with residual hearing through 2000 Hz often do better with hearing aids than with implantation. Computed tomography findings may also preclude implantation. The absence of the cochlea (Michel deformity), and a small internal auditory canal (associated with cochlear nerve atresia) are contraindications to implantation on that side. Other forms of dysplasia are not necessarily contraindications. However, when implantation of a dysplastic cochlea is to be undertaken informed consent is especially important. Cochlear implants in these patients are associated with increased risk of poor result, CSF leak, and meningitis.
The presence of active middle ear disease is a contraindication to surgery. This process should be treated and resolved before implantation. In a study by Luntz otitis-prone patients were treated by protocol (antibiotics, PETs, etc.) before surgery and then implanted (often with PETs in place). No delay was necessary when compared with patients who were not otitis-prone. Several were noted to have inflamed middle ear mucosa on implantation which required removal in order to identify the round window, but did very well with few postoperative episodes of otitis. Children with a history of chronic suppurative otitis media were implanted in a study by El-Kashlan without demonstrable early or late complications. Patients with a history of canal wall down mastoidectomy may need surgery to reconstruct the posterior canal wall or close off the canal before implantation.
Meningitis may lead to hearing loss and ossification of the cochlea. Labyrinthitis ossificans is usually identifiable on CT scan (brightly lit cochlea with obliteration of the basal cochlear duct) and is a relative contraindication when there is a patent contralateral basal turn. MRI is often better at delineating patency of the cochlea and should be pursued if there is any question. Very young children with hearing loss after meningitis should be followed with CT/MRI until they reach implantable age. Early implantation may be indicated if evidence of ossification is noted. Adults and children with acute meningitis should be treated with steroids to avoid hearing loss. Those that do sustain hearing loss secondary to meningitis should be observed for 6 months before implantation due to the substantial number of patients that will regain their hearing in at least one ear. Advanced otosclerosis can also cause ossification of the basal turn of the cochlea. This finding is most often noted on CT scan. This is not a contraindication as long as the surgeon is prepared to perform a drill out or pursue implantation into the scala vestibuli. Patients with otosclerosis can achieve excellent results from implantation.
A diagnosis of neurofibromatosis II (history of progressive hearing loss and suggestive MRI findings), mental retardation, psychosis, organic brain dysfunction, and unrealistic expectations may also be contraindications.
Work-up for a Patient Seeking Cochlear Implantation
- Audiologic examination with binaural amplification
- CT scan/MRI of temporal bones
- Trial of high-powered hearing aids
- Psychological evaluation
- Medical evaluation
- Any workup necessary to discover etiology of hearing loss
The surgical procedure to implant the receiver and electrode array is most often a day-surgery with the patient being discharged shortly after completion of the implantation. Implanting the better or worse-hearing ear is a decision reached by the physician and the patient. The patient should understand the risk of losing all residual hearing in the implanted ear. One recent study by Chen showed no long-term advantage to implanting the patient’s better ear. Thus, many surgeons opt to implant the worse ear and have the patient continue to wear an aid in the best hearing ear.
The surgical procedure begins after the patient receives a general anesthesia. The patient’s head is shaved over the post-auricular area. The extent of hair removal depends on the incision to be used–generally four fingerbreadths above and behind the ear is sufficient. The patient is then prepped and draped in a fashion similar to other otologic procedures. A dummy receiver is placed over the skin and positioned approximately 1 cm posterior to the auricle. A postauricular incision is made 1-2 cm posterior to the implant. Several incisions have been proposed and include a large C-shaped incision, a 4-5 cm superior elliptical extension of the routine postauricular incision, a small 4-5 cm straight incision posterior and an incision posterior-superior to the auricle (minimally invasive procedure introduced by the Cochlear Company). The skin flap is elevated followed by the creation of an anteriorly-based temporoparietal fascia flap. The temporalis musculature and overlying fascia are left intact. A subperiosteal pocket is created medial to the temporalis muscle for placement of the ground electrode. A circular depression is then drilled in the temporal bone cortex superior and posterior to the area to be drilled for access to the round window. Tunnels are often drilled into the surrounding bone in order to place anchor sutures over the receiver.
A complete mastoidectomy is performed with minimal saucerization. A shelf of mastoid cortex can be helpful when securing the array and tucking the excess grounding wire. The facial recess is opened, taking care to avoid injury to the chorda tympani and facial nerves. The round window niche is inspected. A cochleostomy is then drilled over the basal turn of the cochlea just anterior/inferior to the round window. This is carried down to endosteum of the cochlea. The endosteum is then opened using a straight pick. The electrode array is then carefully inserted through the fenestra into the scala tympani. An inserting claw or jeweler’s forceps may be used to advance the electrode array. Excess force should not be used, as the array can easily buckle and cause damage to the internal components. A deep insertion is desired in order to place the electrodes closer to the apex where the highest concentrations of surviving ganglion often are found. A small amount of connective tissue is then packed around the electrode array at the cochleostomy site in order to seal the opening. Care is taken to avoid accidental removal of the array once placed. The ground electrode is tucked into the sub-periosteal pocket and the wound is closed in several layers. No drains are placed. A bulky mastoid dressing is applied. The wound is given several weeks to heal before use of the external processor is attempted. The external processor is held in place by magnetic attraction to the magnet in the implanted receiver.
In patients who have a history of meningitis leading to hearing loss, labyrinthitis ossificans may have caused obliteration of the scala tympani. In this case, the array can be placed into the scala vestibuli. Often the ossification is incomplete and if the surgeon drills forward along the basal coil for 4-5mm the scala tympani will be identified. Care must be taken to avoid injury to the carotid artery which lies just anterior to the cochlea. In some cases of cochlear dysplasia CSF gushers have been encountered. This is managed by allowing the pressured fluid to drain off, and then proceeding with insertion as per routine.
The surgical complication rate after cochlear implantation is estimated to be only 5%. The most common problems are wound infection and wound breakdown. Rarely, extrusion of the device, facial nerve injury, bleeding, CSF leaks and meningitis can occur. Device-related complications include intracochlear damage, slippage of the array, breakage of the implant, and improper or inadequate insertion. Postoperative infection of the surgical site was treated by prolonged courses of postoperative antibiotics by Yu, et al with excellent results. They suggest that a long course of antibiotics and limited I&D will treat the vast majority of wound infections without the need to remove the implant. Those patients that did not respond to this treatment protocol were often found to be immunocompromised. Steenerson reported a 75% incidence of postoperative vertigo, but indicated that these patients did well after undergoing vestibular therapy. Other series do not show as high an incidence, nor the need for vestibular rehabilitation postoperatively. Stimulation of the facial or vestibular nerve by stray electrical current from electrodes outside or near the round window has also been reported. This is usually addressed by “turning off” the responsible electrodes and moving the electrical stimulation to electrodes located within the cochlea.
Recent reports of increased incidence of meningitis in cochlear implant recipients have prompted the CDC to recommend vaccination of implanted or soon to be implanted patients. Children less than 2 years old who have implants should receive pneumococcal conjugate vaccine (Prevnar). Children with implants 2 years and older who have completed the conjugate series should receive one dose of the pneumococcal polysaccharide vaccine (Pneumovax 23 or Pnu-Imune 23). Children with implants between 24 and 59 months who have never received vaccination should receive two doses of pneumococcal conjugate vaccine two months apart and then one dose of pneumococcal polysaccharide vaccine at least two months later. Finally, persons age 5 years and older with cochlear implants should receive one dose of pneumococcal polysaccharide vaccine.
Although the incidence of device failure is very low, occasionally removal of the implant and reimplantation is necessary. These patients do surprisingly well. Alexiades, et al. showed that patients did as well or better after reimplantation (in the same ear) as with their first implant. Thus a history of implantation is not a contraindication to another cochlear implant. Long-term electrical stimulation from a cochlear implant has raised concern for damage to the auditory nerve. However, cochlear implants typically discharge less than one microcoulombs per cm2 of electrode surface and long-term studies have shown no detrimental effects. In fact, studies following patients for up to 13 years show no decline in function. This finding is still true when the study population includes those that have been implanted multiple times.
Research looking at cochlear implants under many environmental strains has shown them to be reliable and safe. Backous, et al showed them to be stable when exposed to extreme barometric pressure changes (as experienced when scuba diving). MRI exposure should be avoided generally, but may be pursued when necessary.
Unless intensive postoperative rehabilitation is undertaken, cochlear implantation is likely to provide little benefit. Each patient’s need for rehabilitation is different based on pre-operative auditory experience. For the prelingually deaf patient, auditory and speech training are imperative if they are to improve their communication abilities. Postlingually deaf patients often need training in more complex listening skills. Cochlear implants in children are successful when the implantation is followed by a intensive treatment by a multidiscipline rehabilitation team. The goal of a pediatric rehabilitation team is to enable the hearing-impaired child to be able to learn passively from his environment. The rehabilitation must address both receptive language skills as well as expressive language abilities. A structured program with dedicated team members is integral to a successful cochlear implant program.
Results of Cochlear Implantation
Cochlear implantation is really the only effective way of treating patients with profound sensorineural hearing loss who do not benefit from hearing aids. Although the perception of a successful implantation might vary from patient to patient, the primary goal of implantation has always been improved speech perception. Since implantation began, physicians have noted a wide range of outcomes. Some patients find little benefit after implantation and may even find the stimulation annoying. Others are able to function normally even without visual cues. Still others are able to listen to and enjoy music. Years of research has given us a better understanding of what variables might influence the results of implantation.
The age of onset of deafness, as well as the length of time since the onset of hearing loss has both been shown to influence outcomes. Several studies have shown that patients who were prelingually deafened show the poorest outcome. Prelingually deaf children implanted before age 6 appear to be able to “catch up” to implanted postlingually deaf children within 2-5 years. These children, like their postlingual counterparts, are able to achieve open-set speech discrimination. Several studies have shown that implantation at an earlier age results in earlier achievement of open-set speech discrimination. Govaerts, et al. showed that 90% of those implanted before age 2 were integrated into mainstream education whereas only 20-30% of those implanted after age 4 were ever integrated. These results are seen in children who are enrolled in aural/oral educational programs and who use oral language as their primary communication modality. The performance of implanted children is far better than those with equal hearing deficits who rely on vibrotactile devices or hearing aids. Generally, implantation of prelingually deaf adolescents and adults is significantly less successful, though results vary widely.
Most authors now believe that the shorter the period of auditory deprivation, the faster and more complete will be the achievement of open-set speech discrimination. This has been shown to be true in the adult population, as well with children. Those patients who are implanted within a short time seem to retain the plasticity of the auditory system better than those who have been deaf for a period of years. Sharma, et al. compared children implanted after different periods of deafness. He showed that children with the shortest amount of time spent without auditory stimuli regained normal cortical responses more rapidly than all others. Specifically, those with 3.5 years of deafness or less showed age-appropriate P1 latencies (a marker of plasticity) after only 6 months of stimulation with a cochlear implant. The length of time required to reach age-appropriate latencies increased with increasing length of auditory deprivation. After age 7 plasticity was greatly reduced.
Waltzman, et al. studied the long-term effects of cochlear implants in children. They followed the children after implantation for five to fifteen years and documented speech perception scores, device extrusion rates, and implant viability. He showed that implantation resulted in significant improvement of patient’s speech perception and that this benefit remained stable (often improving) over the long-term. For the vast majority of his study group this resulted in assimilation into mainstream education. There was no significant incidence of device extrusion or migration and even when device failure necessitated reimplantation, long-term performance was not decreased.
Recent studies looking at the economics of cochlear implants show cochlear implantation improves patient’s quality of life and is cost-effective even in elderly patients (>50 years old). Implantation results in significant benefit to the society as a whole, and to the individual. Unfortunately, cochlear implantation is more often a money-losing effort for everyone involved with implantation. Hospitals and physicians, as well as the other members of the rehabilitation team often find themselves without funding and support.
Partial insertion cochlear implantation has been proposed as treatment for those patients who have residual low-frequency hearing with high-frequency sensorineural hearing loss. The speech processor is coupled with a hearing aid and thus provides maximal aided hearing. The risk of such surgery is loss of the remaining hearing in that ear. Other implant strategies include brainstem implantation for those without an intact cochlear nerve. Good results have been reported.
Nucleus products now come equipped for intraoperative testing. This allows the audiologist to map the patient’s electrode array while the patient is asleep. This is especially useful for infants and small children who are often not cooperative with conventional mapping techniques. Intraoperative repositioning is also a possibility if the mapping shows poor responses.
Bilateral implantation with the possibility of binaural hearing is currently being studied. Gantz showed that most patients with bilateral implants perform better on sound localization, but only some do better with auditory performance in noise (at one year). Despite this, most studies report increased satisfaction with two implants. At least one company is proposing a system with only one processor and receiver but implanted electrodes in both ears. Despite these results, the cost of a second implant is prohibitive. Summerfield argued that the quality of life likely to be gained (across society as a whole) by unilateral implantation is higher, per unit of expenditure, than with bilateral implants.
Implantation for patients with asymmetric sensorineural hearing loss may soon be approved. Cochlear implants are expected to help these patients with sound localization and speech comprehension in noise.
The Cochlear Company is currently testing a new “Softip” electrode array which is advanced off a stylet after traversing the straight section of the basal turn. The array then curls around the modiolus until fully inserted. The technique has been shown to cause less trauma to the basilar membrane and intracochlear structures in preliminary studies. The Cochlear Co. is also marketing its new “minimally invasive” approach which allows for implantation through a small (4-5cm) incision over the post auricular area which does not require shaving of the hair in that area. Warm response is reported by those performing this approach.
Cochlear implantation is no longer experimental. It is the treatment of choice for children and adults with severe-to-profound hearing loss. Significant gains in open-set speech recognition have been demonstrated by most of those who undergo implantation. Early implantation, whether in pre or postlingual patients, has shown to be effective at moving an otherwise marginalized segment of society into the mainstream. Implantation is cost-effective and results in high patient satisfaction. Although cochlear implants are still a rough and awkward imitation of our natural sense, they offer hope to thousands who must otherwise live in a silent world.
CONGENITAL AURAL ATRESIA
Congenital aural atresia is a term used to refer to a spectrum of ear deformities present at birth that involve some degree of failure of the development of the external auditory canal (EAC). Often, the malformation will also involve the tympanic membrane, ossicles and middle ear space to varying degrees. While associated abnormalities of the auricle are common, the inner ear development of these patients is most often normal. The challenge to the otologist is to restore the sound conduction pathway through the atretic EAC and malformed middle ear to the normal cochlea.
The incidence of congenital aural atresia is approximately 1 in 10,000 to 20,000 live births. Unilateral atresia occurs three to five times more commonly than bilateral atresia. Males are more often affected than females and in unilateral cases the right ear is more commonly involved. This anomaly most often occurs sporadically although cases of autosomal dominant or recessive inheritance have been reported. Aural atresia has been reported to occur in association with hydrocephalus, posterior cranial hypoplasia, hemifacial microsomia, cleft palate and genitourinary abnormalities. It has also been described as part of various syndromal abnormalities including Treacher-Collins, Goldenhar’s, Crouzon’s, Mobius’, Klippel-Feil, Fanconi’s, DiGeorge, VATER, CHARGE and Pierre Robin.
Congenital aural atresia occurs as a result of abnormal development of the first and second branchial arches and the first branchial groove. Development of the external ear begins during the fourth week of gestation as six mesenchymal proliferations enlarge to form ridges known as the hillocks of His. These hillocks, which surround the first branchial groove or primitive meatus, fuse to form the primitive auricle by the third month of gestation.
The EAC develops from the first branchial groove beginning in the eighth week of gestation. Epithelial cells from the meatus proliferate forming a solid core of cells, known as the meatal plug. This core of cells migrates medially toward the outgrowth of the first branchial pouch, which will eventually form the middle ear cleft. The meatal plug contacts the middle ear cleft by the ninth week of gestation. This solid core will then recanalize to form the epithelial lined EAC, but not until the sixth or seventh month of gestation.
Ossicular development begins in the fourth week and, at this time, the malleus and incus appear as a fused mass. Separation into two distinct ossicles typically occurs by the eighth week of gestation. The first branchial arch, Meckel’s cartilage, contributes to the development of the head and neck of the malleus and the body and short process of the incus. The second branchial arch, Reichert’s cartilage, leads to the development of the manubrium of the malleus, the long process of the incus and the stapes suprastructure. By the sixteenth week of gestation, the ossicles are of adult size.
Inner ear development from the otic placode begins during the third week of gestation. Invagination of the otic placode to form the otic vesicle is apparent by week four, and by the sixth week the semicircular canals have taken shape. The utricle and saccule have formed by the eighth week. Development of the cochlea begins during the seventh week, and by week twelve the complete two and a half turns have formed. The membranous labyrinth is entirely developed by fifteen weeks gestation and ossification of the surrounding otic capsule is complete by twenty-three weeks gestation.
The nerve of the second branchial arch is the facial nerve. Its development begins with the differentiation of neuroblasts from the acoustico-facial primordium between four and five weeks gestation. The course of the nerve is completely formed by seventeen weeks. However at this time, the nerve is located in a more anterosuperior position. The eventual migration of the facial nerve to its normal adult position is dependent upon the normal development of the tympanic ring and mastoid.
From this summary of otologic embryology, we can see that completion of development of the external ear occurs rather early in gestation while the recanalization of the EAC occurs later. Therefore, a severely deformed auricle is likely an indicator of associated EAC, middle ear, facial nerve and possibly inner ear anomalies. In contrast, aural atresia in the presence of a normal auricle most likely represents a later arrest in development and has a higher likelihood of normal middle and inner ear structures.
Several different classification or grading systems for congenital aural atresia are present in the otolaryngology literature. Altmann’s classification, first reported in 1955 but still widely utilized today, divides atresia into three groups based on the clinical assessment of the severity of the malformation. Group I atresia is characterized by a small EAC, hypoplastic temporal bone and tympanic membrane (TM), a normal or small middle ear cleft and normal or mildly deformed ossicles. Group II includes those cases with an absent EAC, an atretic plate, a small middle ear space and fixed and malformed ossicles. Group III is characterized by an absent EAC, a severely contracted or absent middle ear space, and absent or severely malformed ossicles.
De la Cruz made modifications to Altmann’s classification system so that cases are categorized into major and minor malformations. The minor category is characterized by normal mastoid pneumatization, normal oval window, reasonable oval window-facial nerve relationship and a normal inner ear. The major category is comprised of cases with poor pneumatization, abnormal or absent oval window, abnormal course of the horizontal facial nerve and inner ear anomalies.
Yet another classification system was introduced by Schuknecht in 1989. This system divides atresia cases into four types based primarily on intraoperative findings and the type of surgical repair required. Type A atresia is limited to the cartilaginous EAC and is addressed with meatoplasty. Type B atresia is characterized by narrowing of both the cartilaginous and bony EAC along with a small TM and mild deformity of the malleus and incus. This type of atresia most often requires canalplasty, possibly with ossicular chain reconstruction (OCR) as well. Type C atresia cases have complete EAC atresia but a well-pneumatized middle ear and mastoid. The TM and ossicular malformations are more severe than in type B and there is a higher likelihood of facial nerve anomalies. Canalplasty and OCR will be necessary to correct type C atresia. Type D atresia involves complete EAC atresia and poor middle ear pneumatization. In these cases, associated facial nerve or inner ear anomalies often preclude surgical intervention.
Probably the most clinically useful classification system was introduced by Jahrsdoerfer in 1992. This system establishes a score (up to 10) based upon findings of high resolution CT scans of the temporal bone. The parameters of an open oval window, width of the middle ear cleft, facial nerve course, malleus-incus complex, mastoid pneumatization, incudostapedial continuity, round window patency and auricle appearance are assigned a value of one point. The presence of a stapes is given a higher priority and assigned two points. The final score has been used to predict the likelihood of successful atresia surgery. A score of 8 out of 10 correlates to an 80% chance for restoration of hearing to normal or near-normal levels defined as speech reception thresholds (SRT) between 15 and 25 dB. Cases with a score of less than or equal to 5 are generally not considered for surgical intervention.
The first step in the evaluation of a patient with congenital aural atresia is to obtain a complete history and perform a thorough physical exam. Given that the majority of these cases will be discovered in the newborn, the history is focused on the details of pregnancy. It is important to ask about prenatal care and to determine if the mother was exposed to infections, drugs or alcohol during pregnancy. Parents should also be questioned about any family history of ear deformities or syndromal anomalies. Physical examination will, of course, involve a complete head and neck exam. Specifically, the degree of microtia is assessed and the severity of EAC atresia noted. In cases of EAC stenosis, attempts should be made to visualize the TM and ossicles and their presence and position documented. Additionally, examination of overall craniofacial development is necessary to assess for the presence of associated branchial arch anomalies.
Next, accurate audiologic evaluation in the newborn period is mandatory. Infants with any degree of ear anomaly should be marked as high-risk for associated hearing loss and have auditory brainstem response (ABR) testing before leaving the hospital. The initial priority in cases of unilateral atresia is to evaluate the auditory function of the unaffected ear. Normal hearing in one ear will allow for essentially normal speech and language development. However, the incidence of both conductive and sensorineural hearing loss in the nonatretic ear is greater in patients with unilateral atresia than in the general population. Therefore, it is essential that any auditory dysfunction in the “normal” ear be diagnosed early on so that appropriate amplification can be implemented if necessary. If surgical repair of unilateral atresia is considered later on, audiologic testing to confirm normal cochlear function in the involved ear will be necessary.
In cases of bilateral aural atresia, early evaluation with both air and bone conduction ABR testing is necessary. The bilateral conductive component makes testing somewhat more difficult by creating a masking dilemma. Evaluation of ear specific cochlear function is possible by measuring the Wave I response ipsilateral to the stimulation. Patients with bilateral atresia should be fitted with bone conduction hearing aids as early as possible to optimize speech and language development.
Once auditory function has been established, either via the unaffected ear in unilateral atresia cases or with bone conduction aids in bilateral cases, further evaluation is not necessary until the child has reached the age of 5 or 6 years. At this time, a high resolution CT scan of the temporal bones in both the axial and coronal planes is indicated. This study will allow evaluation of middle ear pneumatization, ossicular anatomy, inner ear morphology and the course of the facial nerve. This is the most useful study to decide if a patient is a candidate for surgery and to predict the likelihood of successfully reestablishing normal hearing as discussed previously. An indication for obtaining a CT scan earlier would be those patients with congenital aural atresia that present with a draining ear or acute facial palsy, which could indicate an underlying cholesteatoma.
There are two absolute requirements for a patient with congenital aural atresia to be eligible for surgery: 1) normal inner ear morphology demonstrated on CT scan and 2) normal cochlear function demonstrated by audiologic testing. A score of 5 out of 10 or less by the CT scan grading system may be considered a contraindication to atresia surgery because these patients are not likely to have an appreciable hearing improvement and are at higher risk for surgical complications. Patients with a score of 6/10 are considered “marginal” candidates, 7/10 “fair”, 8/10 “good”, 9/10 “very good” and 10/10 “excellent.”
Once the decision to operate has been made, the timing of repair must be planned. Patients with auricular malformation should undergo microtia repair first to avoid scar tissue formation compromising the local blood supply. Most authors agree that this process should begin around age 5 to 6 years at which time costal cartilage is sufficiently developed for harvest and optimal development of the mastoid process has occurred. Opinions vary, however, as to whether atresia repair should be performed between Stages 2 and 3 of microtia repair or 2 months after the final stage of microtia repair.
There are basically two techniques for atresia repair—the transmastoid approach and the anterior approach. According to most authors, the transmastoid approach is not used as frequently but may be a preferable option in some cases. This approach begins with drilling the mastoid to allow identification of the sinodural angle, which is then followed anteriorly to the antrum. The lateral semicircular canal is identified and used as a landmark. The facial recess is opened and the incudostapedial joint (if present) may be separated. The atretic plate is then carefully removed. Ossiculoplasty and tympanoplasty proceed in the usual fashion and the newly created EAC is lined with a split thickness skin graft. The main disadvantages of the transmastoid approach are the creation of a larger defect that must be skin grafted, prolonged healing of the cavity and the presence of a mastoid bowl that requires lifelong maintenance.
The anterior approach, popularized by Jahrsdoerfer, is the most common method of atresia repair utilized today. In this technique, drilling begins at the atretic plate just posterior to the temporomandibular joint and inferior to middle fossa dura. Dissection proceeds medially following these two landmarks to the epitympanum where the fused malleus and incus can be identified. The safest area to drill is anterosuperiorly because the facial nerve is consistently located medial to the ossicles in the epitympanum. The most likely area to encounter an aberrant facial nerve is while drilling posteroinferiorly so this should be performed only after identification of other landmarks. Care must be taken not to drill directly on the ossicular mass to avoid trauma to the inner ear. The atresia plate is thinned and removed and any fixation of the ossicles to the atretic bone is lysed, either sharply, or, perhaps more safely, with the carbon dioxide laser.
Next, continuity of the ossicular chain must be assessed and if found to be intact, no manipulation is indicated. Cases in which the incudostapedial joint consists of just a fibrous attachment or is altogether absent warrant ossiculoplasty with a PORP. Unstable or absent suprastructure of the stapes requires reconstruction with a TORP. Both of these statements assume a mobile stapes footplate, which is most often the case. In those rare patients with a fixed footplate, some type of fenestration procedure will be necessary along with ossiculoplasty.
A temporalis fascia graft that had been harvested earlier and allowed to dry is used to recreate the TM. Optimally, the new TM will be centered on the ossicular mass to maximize hearing results. If a PORP or TORP had been used for ossiculoplasty, placing a small piece of cartilage between the table of the prosthesis and the fascial graft will help to minimize chances of extrusion. Finally, a split thickness skin graft, .012-.015 inches in thickness and approximately 6x6cm, is harvested for lining the new EAC. Most authors report using a donor site of the upper inner arm, but alternatives include the upper thigh or buttock. The skin graft is placed into the EAC, overlapping the TM facial graft medially. Care must be taken to ensure that all bone is covered with skin and that the skin graft edges are not folded over on itself. The ear canal is then packed with Nu-gauze or Merocel sponges impregnated with antibiotic ointment.
The native or reconstructed auricle is often located anteroinferior in relation to the newly created EAC. The auricle can be repositioned by undermining soft tissue and possibly excising redundant postauricular skin. An external meatus is then created by excising skin, subcutaneous tissue and cartilage from the auricle. Alternatively, an anteriorly based flap of conchal skin and cartilage can be incised and folded into the new EAC to line the anterior portion of the canal. The lateral edge of the skin graft is brought through the meatus and sutured to the skin edges and this area is packed similarly to the medial EAC. The postauricular incision is closed in the usual fashion, possibly with the addition of some tacking sutures to the periosteum to maintain the posterior position of the auricle and to keep the meatus widely open. A mastoid dressing is applied.
The mastoid dressing is removed on postoperative day 1. The timing for removal of the EAC packing varies between authors, some recommend removing it altogether at 10 days to 2 weeks. Other recommend removing and replacing the meatal packing at 2 weeks, starting antibiotic ear drops at that time then removing the entire pack at 3 weeks. Frequent visits are required after packing removal to address any granulation tissue formation and to remove any desquamated skin.
The hearing results reported in the literature after surgery for congenital aural atresia are somewhat difficult to interpret because of different classification systems used to describe the atresia preoperatively, different criteria for selecting surgical candidates, different definitions of a “successful” outcome and different time periods of patient follow-up after surgery. However, the majority of authors consider a successful hearing result to be 25-30dB or less after surgery. Using this definition, the percentage of cases with a successful outcome reported in the literature varies from 12-71%.
Unfortunately, the majority of these papers are citing the hearing results obtained in the early postoperative period. An important concept to be familiar with when counseling patients about this surgery preoperatively is the stability of the hearing results over time. This issue was addressed by Lambert in 1998 by comparing early postoperative hearing levels (<1yr postop) to levels after longer follow-up (1-7.5yrs, avg 2.8yrs). He found that 60% of cases had hearing levels of 25dB or better and 70% were at 30dB or better in the early postoperative period. This diminished to 46% and 50% with longer follow-up. Additionally, he found that nearly one third of cases required revision surgery, most often for restenosis of the EAC or lateralization of the TM. After revision surgery, hearing levels of 25dB or less were achieved in 50% of cases and levels of 30dB or less in nearly two thirds of cases. He also commented that of those patients with an exceptional result after primary surgery (hearing level 10-20dB) 83% maintained this outcome over longer periods of follow-up.
The most frightening complications of atresia surgery are facial nerve injury and iatrogenic hearing loss, however canal restenosis, TM lateralization and chronic infection are much more common. The rate of EAC restenosis is also variable in the literature, ranging from 8-50% of cases. These figures have been found to correlate to the initial severity of the atresia with more severe cases having a higher likelihood of restenosis. If the restenosis is mild a recurrence of the conductive hearing loss is unlikely and the patient may be observed. More significant narrowing will cause epithelial trapping and predispose the patient to infection; these cases typically will require dilation or revision canalplasty. If restenosis is limited to the lateral soft tissue portion of the EAC and is caught early it may be responsive to injection with steroid solution.
A lateralized TM presents with a gradual worsening of the hearing level and can occur up to 12 months after surgery. This problem has been reported to occur in 5-26% of cases and is best prevented by meticulous technique at the primary surgery. Taking care to create a good bony annulus upon which to anchor the graft, anchoring the graft medial to the malleus as well as using a silastic button to hold the fascial graft in place will all help to prevent TM lateralization.
Chronic infection of the newly constructed EAC may occur as a result of the lack of normal keratin migration in the skin-grafted canal and the lack of production of protective cerumen. Buildup of keratin debris and trapping of water can lead to epithelialitis and chronic otorrhea. This problem can be minimized by creating a widely patent meatus and addressing any canal restenosis. Patients should be counseled on aural hygiene and recommended to return for microscopic debridement once or twice a year.
The facial nerve follows an aberrant course in 25-30% of cases of congenital aural atresia. It is typically anterolaterally displaced in comparison to its normal course. The bend at the second genu tends to be more acute and the nerve crossed the middle ear in a medial-to-lateral direction so that at the level of the round window, the nerve may be lateral to the middle ear space and encased in atretic bone. Most reports in the literature say that in the hands of experienced surgeons the incidence of facial nerve injury is 1.0-1.5%. The majority of injuries reported were temporary palsies that had return to normal function over several months. Facial nerve transection and permanent dysfunction is exceedingly rare. In a review of over 1,000 surgeries for congenital aural atresia, Jahrsdoerfer and Lambert report only 10 cases of facial nerve injury. In their review, they noted five situations in which the facial nerve was most susceptible to injury: making the skin incision; dissecting in the glenoid fossa; during the canalplasty; transposing the facial nerve; and dissecting soft tissue in the preauricular area. Careful study of the preoperative temporal bone CT with mapping of the facial nerve course is of paramount importance to avoid this devastating complication. Intraoperative facial nerve monitoring and improved imaging techniques both help to avoid nerve injury.
High frequency sensorineural hearing loss has been reported to occur in up to 15% of patients undergoing atresia surgery. This occurs either as a result of transmission of drill energy to the inner ear while removing atretic bone, direct drill injury to the ossicles or traumatic manipulation of the ossicles. Although the first mechanism is largely unavoidable, the others can be prevented by using meticulous surgical technique around the ossicles. Although the hearing loss generally occurs in the 4,000-8,000 Hz range and does not affect the speech frequencies, cases that have experienced a loss in speech discrimination have been described.
The management of the patient with unilateral congenital aural atresia remains controversial in today’s literature. Historically, the teaching has been that a patient with normal hearing in one ear is capable of normal speech and language development. This idea, along with the potential complications of atresia surgery and the unpredictability of hearing results after surgery, have lead many authors in the past to recommend against surgery in unilateral atresia cases. However, recent research has shown that children with unilateral hearing losses do suffer from auditory, linguistic and cognitive impairments that can have a negative effect on their education. Additionally, a unilateral hearing loss causes difficulty with sound localizaton and speech recognition in the presence of background noise. These impairments, along with improved methods of patient selection using high resolution CT and improved surgical techniques with more predictable hearing results after surgery, have lead some authors to begin recommending surgical repair of unilateral atresia.
If surgery is elected for the patient with unilateral atresia, the second issue of controversy is the timing of repair. Many authors would recommend postponing any surgery until the patient is old enough to participate in the decision making process and understand the potential complications and limitations of repair. More recently, several authors are recommending unilateral atresia repair at the age of 5-6 years after microtia repair, as in bilateral cases. These authors emphasize the importance of appropriate patient selection for unilateral repair, primarily reserving surgery for those patients that have favorable anatomical findings on CT scan and whose parents have realistic expectations for surgery and understand the demands of postoperative follow-up.
Nearly every report on congenital aural atresia begins by saying that it is among the most difficult and challenging surgeries for the otologic surgeon. That being said, in the hands of experienced otologists, repair of this deformity can be performed safely and with predictable results. The goals of atresia surgery are to restore functional hearing, preferably without the requirement of a hearing aid, and to reconstruct a patent, infection-free external auditory canal. Successful accomplishment of these goals, in the face of such an operative challenge, can make atresia repair one of the most rewarding surgeries for the otologic surgeon.
INFECTIONS OF THE EXTERNAL EAR
Infections of the External Ear
Anatomy and Physiology
The external ear is an area commonly subjected to acute and chronic inflammatory conditions. It consists of the auricle and external auditory meatus. The auricle is mostly composed of fibroelastic cartilage to which the skin and a small portion of subcutaneous tissue are closely attached, except in the lobule where there is fat and no cartilage. The external auditory meatus (EAM) is a skin-lined canal approximately 2.5 cm in length and ends medially at the tympanic membrane. The lateral cartilaginous portion comprises approximately 40% of its entire length, while the remaining 60% is osseous, formed primarily by the tympanic ring. The dehiscences in the anterior wall of the cartilaginous portion of the canal are known as the fissures of Santorini. They may allow spread of infection from the canal lumen into the preauricular soft tissues, parotid gland, and temporomandibular joint. Because of the oblique position of the tympanic membrane, the posterosuperior portion of the canal is 6 mm shorter than the anteroinferior portion. The canal is S-shaped, curving slightly superiorly and posteriorly from lateral to medial. The narrowest portion of the canal is at the junction of the cartilaginous and bony portions, termed the isthmus. Medial to the isthmus, the canal courses inferiorly and ends in the inferior tympanic recess.
The EAM is related to contiguous structures on all but its lateral surface. Medially, it is bound by the tympanic membrane, which when intact is a good barrier to the spread of infection. Superiorly, it is separated from the cranial fossa by a thick plate of bone, which usually prevents direct intracranial extension of infection. Posteriorly, the bony canal abuts the mastoid cavity. Several vessels penetrate the canal, which may be involved in the hematogenous extension of infection to the mastoid segment. Posterior to the cartilaginous canal, there is dense connective tissue overlying the mastoid portion of the temporal bone, which may become secondarily infected. Anteriorly, the canal is related to the glenoid fossa of the TMJ and the parotid gland. Inferiorly, the canal is related to the infratemporal fossa.
The external ear is innervated by contributions from the trigeminal, facial, glossopharyngeal, and vagal nerves as well as from the cervical plexus (greater auricular nerve). It receives its arterial blood supply from the superficial temporal and posterior auricular branches of the external carotid artery. The deep auricular branch of the internal maxillary artery serves the more medial canal and lateral TM. Venous drainage is via the superficial temporal and posterior auricular veins. The posterior auricular usually drains into the external jugular vein, but may also drain to the sigmoid sinus through the mastoid emissary vein. The lymphatic drainage of the canal is important with regard to the spread of infection and cancer. Inferiorly, the canal drains into the infra-auricular nodes posterior to the angle of the mandible. The anterosuperior canal empties into the preauricular nodes of the parotid and superior deep cervical nodes.
Posteriorly, the lymphatics drain to postauricular and superior deep cervical nodes. Finally, the lymphatics of the antihelix and concha empty into the nodes along the apex of the mastoid process, whereas those of the superior part of the auricle drain into postauricular nodes.
The entire EAM is lined with squamous epithelium, which is thicker in the cartilaginous portion (0.5 to 1mm) than the osseous portion (0.2mm). In the cartilaginous canal, the skin contains sebaceous and apocrine glands with many hair follicles. Together, the hair follicle, sebaceous gland, and apocrine gland are termed the apopilosebaceous unit. Invagination of the epidermis forms the outer wall of the hair follicle. The space between this outer wall and the hair shaft is termed the follicular canal. The excretory ducts of the sebaceous and apocrine alveoli drain into these follicular canals. In the normal ear, the secretions of these glands, combined with the desquamated keratin layer from the stratum corneum, form a water-repellant, acidic, waxy coat of cerumen that serves as a barrier against infection and injury to the skin. Motion of the ear canal provided by ordinary chewing movements together with the process of epithelial proliferation and lateral migration propel the cerumen outward in a self-cleansing manner.
Compromise of any the protective features of the canal can lead to colonization and invasion by pathogenic organisms. Obstruction of the drainage of the glands into the follicular canal can occur in response to increased temperature and humidity within the canal. Absorption of moisture by the stratum corneum leads to hyperhydration and maceration of tissue within the canal, which gives the patient an uncomfortable sense of fullness and itching. Any response that leads to trauma to the canal skin, such as instrumentation, excessive cleansing or scratching, allows for invasion of exogenous or endogenous organisms through breaks in the skin.
Otitis Externa is an infection of the external auditory canal (EAC) that can be divided according to the time course of the infection: acute, subacute, or chronic. Acute otitis externa (AOE) is a bacterial infection of the EAC, commonly referred to as “swimmer’s ear” that can further be divided into preinflammatory and acute inflammatory stages. The acute inflammatory stage may be mild, moderate, or severe. The preinflammatory stage begins with itching, edema and a full sensation in the ear. As the infection progresses increased itching and pain ensues with mild erythema and edema on physical exam, however the canal lumen remains patent with cloudy secretions. During the moderate phase, the itching and pain intensify, and although the lumen remains patent, significant edema and debris decrease its size. Secretions are noted to be exudative and more profuse. Finally, in the severe stage of the disease, the pain is usually intolerable and is often intensified by manipulation of the skin and soft tissue around the ear. The lumen of the EAC may be obliterated by edema, debris and purulent otorrhea. The auricle and periauricular soft tissues are often involved, and regional lymph nodes may become palpable. In patients where the disease does not resolve after treatment, a subacute or chronic form may occur. This condition can be described as a spectrum of disease ranging from mild drying and scaling of the canal skin to complete obliteration of the lumen by the chronically infected and hypertrophic skin.
The most commonly isolated pathogens are Pseudomonas aeruginosa and Staphylococcus aureus. Other pathogens less commonly cultured include Proteus mirabilis, Streptococci species, coagulase negative Staphylococci, and various gram negative bacilli. The treatment of otitis externa involves a strategy intended to resolve the infection while promoting the restoration of the external auditory canal to its original healthy state. Four fundamental principles predominate and include: 1) frequent and through atraumatic cleansing of the canal through careful suctioning and debridement under microscopy (may need to be repeated frequently depending on severity), 2) use of the appropriate topical antibiotics (insertion of an otowick may be necessary to facilitate application of drops medially), 3) treatment of associated inflammation and pain, 4) and recommendations for prevention of future infections (i.e. dry ear precautions).
In most cases of uncomplicated AOE, topical antibiotics are the first-line treatment choice. There is no evidence that systemic antibiotics alone or combined with topical preparations improve treatment outcome over topical antibiotics alone. However, serious manifestations of the disease, such as periauricular cellulitis, necessitate the use of systemic antibiotics based on culture sensitivities. When the status of the TM is unknown, the ototoxic potential of topical antibiotics must be considered. The risk of ototoxicity by ototopical preparations has been debated for years. Estimates vary widely from 0.01% to 3%. Currently, the only ototopical drop approved by the FDA for use in an open middle ear is ofloxacin.
Chronic Otitis Externa
Chronic otitis externa (COE) is an inflammatory process of the ear canal due to bacterial, fungal, or other dermatologic disorders. COE can result from recurrent otitis externa, chronic purulent otitis media with perforation, or eczematoid dermatitis. The disease can be defined by having persistent symptoms for more than 2 months, which include unrelenting pruritus, mild discomfort, and dry flaky skin in the EAC. On exam, the EAC skin exhibits asteatosis (lack of cerumen), dryness and hypertrophy. Partial canal stenosis from the hypertrophied skin is common. Mucopurulent otorrhea is occasionally found. Culture reports vary widely and are often distorted because patients have been prescribed various antibiotics before referral to an otolaryngologist. One study reports S. aureus, Pseudomonas, and fungi as the most predominate pathogens. Management is similar to that of AOE. Multiagent topical treatments and frequent cleanings are often necessary. Topical steroid cream may help alleviate the chronic itching and resultant excoriations often present with this condition. Rarely is surgical intervention necessary. However, if medical management fails, surgical procedures to enlarge and resurface the EAC, such as conchal meatoplasty, are indicated.
Furunculosis (Acute localized otitis externa)
Furunculosis is a localized infection, usually found in the lateral one third of the posterosuperior aspect of the EAM that results from obstructed apopilosebaceous units. The medial canal is often normal in appearance and to palpation. The most common pathogen is Staphylococcus aureus. The primary lesion is often a small pustule that may enlarge to become a furuncle. The symptoms include localized pain and itching, and may include pain with mastication if the lesion involves the anterior wall. If the lesion occludes the canal, hearing loss may be present. Upon examination, edema, erythema, tenderness and occasionally fluctuance are present. Limited lesions that have not progressed to form an abscess are treated with local heat, analgesics and oral anti-staphylococcal antibiotics. If spontaneous drainage does not occur, and the lesion progresses into an abscess with cellulitis, incision and drainage after local anesthesia is indicated. Extension of the infection to the pinna and periauricular soft tissues may warrant parenteral antibiotic therapy.
Otomycosis is a fungal infection of the skin of the EAC. Fungi can be either the primary pathogen or superimposed on bacterial infections. Many fungi have been implicated in the disease process, however the most common organisms isolated are Aspergillus and Candida. The initial symptoms of fungal otitis externa are often indistinguishable from bacterial OE. The most common symptoms of otomycosis are pruritis deep within the ear and an irresistible urge to scratch. The itching generally progresses to dull pain with or without drainage. The accumulation of fungal debris in the inflamed, narrowed canal often leads to a complaint of hearing loss. Tinnitus is also a common presenting complaint. Physical examination generally demonstrates canal erythema, mild edema and the presence of white, gray, or black fungal debris within the canal. Treatment is directed at thorough cleaning and drying of the canal followed by the application of topical antifungal medication. There are currently over 20 topical agents recommended for the treatment of otomycosis, however, controversy exists as to the first-line agent of choice. Lucente describes an effective regimen as: 1) clean the canal thoroughly and dry completely, 2) apply topical Cresylate for 5 minutes, 3) flush the canal gently with Domeboro solution and dry again, 3) apply a thin coating of nystatin-triamcinolone ointment (Mycolog II) throughout the length of the canal under microscopy, 5) give the patient a prescription for Mycolog II to use at home once daily. Occasionally, Lucente describes having to use systemic antifungal agents in conjunction with intensive topical therapy in patients with refractory diseases. Should the ear become macerated and wet topical powders are preferable. In patients with previous mastoid surgery, Gentian violet is well tolerated for fungal infections involving the mastoid cavity.
Granular Myringitis is the result of localized chronic inflammation of the lateral surface of the pars tensa of the tympanic membrane and is characterized by persistent, incompletely epithelialized granulation tissue over the involved area. It is a poorly understood entity with few cases documented in the literature, which is why its incidence is difficult to estimate. Toynbee was the first to record a description of granular myringitis in 1860, when he noted a case of “catarrhal inflammation of the dermoid layer after measles” with a “polypoid growth from the surface, especially posteriorly”. It has been reported to occur as a result of primary acute myringitis, a sequela of a previous OE, or a perforation of the TM. Gram –negative bacilli are the most commonly cultured organisms, especially Pseudomonas and Proteus species, however, there is no evidence that any one type of bacteria or fungi is associated with this disease.
The course is generally chronic with inflammation confined to the outer epithelial and underlying fibrous middle layers of the TM. The layers become replaced by granulation tissue, which may extend over the entire surface of the eardrum, if neglected. The usual presenting complaint is a foul smelling discharge from one ear, although many patients remain asymptomatic. Other common complaints are slight irritation and fullness in the ear without significant pain or hearing impairment. On physical examination, the TM is usually obscured by a mucopurulent discharge with “peeping” granulations. There is no perforation of the TM to be found with granular myringitis, which distinguishes it from chronic suppurative otitis media with perforation.
Treatment includes careful and frequent debridement of the ear with the application of anti-Pseudomonal antibiotics, occasionally combined with steroids for at least two weeks. If no resolution of the granulation tissue occurs after topical treatment, some form of topical chemical destruction of the granulation tissue should be used without destroying the underlying fibrous middle layer of the TM. Any chemical agent should be left in contact for less than 2 minutes and only applied once a week in order to avoid necrosis and perforation of the fibrous layer. One example, described by Yinglin, is a 0.5% solution of formalin. Other agents include 50% chromic acid, ferric perchloride, solid silver nitrate, trichloracetic acid, and pure carbolic acid.
This is a form of viral involvement often confined to the tympanic membrane and primarily involves younger children. The presenting symptom is one of severe pain without fever and hearing loss. Upon examination of the ear, the inflammation is limited to the TM and adjacent canal wall and appears as multiple blebs that are reddened and inflamed. The vesicles are usually hemorrhagic and when ruptured produce a significant amount of bloody otorrhea. Unless there is a secondary bacterial invasion, the middle ear is not involved. The condition is self-limiting with resolution in 2 to 3 days. Treatment is aimed at pain relief and systemic and topical antibiotics to prevent secondary bacterial infection. Incision of the blebs is not recommended, due to the possibility of secondary infection, as this does not appear to change the rate of recovery.
Necrotizing External Otitis (Malignant Otitis Externa)
Necrotizing External Otitis (NEO) is a potentially lethal infection of the EAC and adjacent structures typically seen in elderly diabetic or immunocompromised patients. Pseudomonas aeruginosa is the bacteria most commonly responsible for this infection, which begins as an acute otitis externa and frequently progresses to a skull base osteomyelitis with resultant cranial neuropathies. Meltzer and Kelemen first described the disease process in 1959, but the name is credited to Chandler with his precise description of the clinical entity in 1968. The diagnosis of NEO is based on clinical and laboratory evidence along with the suspicions of the treating physician.
The typical patient is an elderly diabetic with poor metabolic control and evidence of otitis externa not responding to the usual local therapy. The typical complaints are deep-seated aural pain, discharge, and fullness. A history of diabetes or an immunocompromised state (neoplasm, immunosuppressive therapy, HIV, etc.) should be elicited. Examination of the involved ear canal reveals inflammation and granulation tissue at the bony cartilaginous junction. Purulent secretions are common, and excessive inflammation may occlude the canal and obscure the TM. Disease beyond the EAC may extend anteriorly into the parotid through the fissures of Santorini or inferiorly into the soft tissue below the tympanic ring. Cranial nerve involvement may appear as early as one week after the onset of symptoms, with the facial nerve most commonly involved, followed by X and XI. Various imaging techniques have been employed to help is the
diagnosis of NEO, including plain films, computerized tomography (CT), technetium-99(Tc99) bone scan, gallium scan, and magnetic resonance imaging. Computerized tomography scanning is particularly useful for following soft tissue extension of infection and subtle bony changes, and is the radiological test of choice today. Tc99 scanning and gallium scans are reliable in identifying osteomyelitis of the temporal bone and skull base. The gallium scan reverts to normal with successful treatment, and is therefore useful for evaluating effectiveness of therapy.
Cohen and Friedman established diagnostic criteria to distinguish NEO from AEO, based on obligatory and occasional signs. The signs were determined from a review of the current literature and were divided into major signs (appeared in 100% of cases) and minor signs (appeared only in some of cases). Major signs included: pain, exudates, edema, granulations, microabscess, positive Tc99 scan, and failure of local treatment after more than 1 week. Minor signs included: Pseudomonas, positive radiograph, diabetes, cranial nerve involvement, debilitating condition, and old age. It was noted, however, that Pseudomonas was found in 98% of cases reported, but did not technically meet the requirement for being a major sign of NEO.
Treatment with parenteral anti-Psuedomonal antibiotics should be continued for a minimum of 4 weeks. Local canal debridement is an essential part of therapy and should be started immediately and continued until granulation tissue resolves and healing ensues. Pain control is usually necessary and, underlying disease states must be controlled. The use of topical antimicrobial agents is controversial, because they are insufficient for invasive infection and tend to hinder culture isolation of the offending pathogen. Hyperbaric oxygen has been used with varying success in some reports. Resolution of otalgia, decreased drainage and a falling ESR indicate a response to therapy. The duration of antimicrobial therapy depends on serial gallium scans performed at 4-week intervals. Surgical debridement of tissue and infected bone is usually reserved for those patients not responding to medical management. There is no universal agreement on the need for prophylactic or therapeutic facial nerve decompression.
Mortality remains significant with the death rate essentially unchanged n 20 years despite the introduction of newer antibiotics. Increased mortality is associated with mental status deterioration and cranial nerve involvement, with the highest mortality seen in cranial polyneuropathies. Recurrence is not uncommon with rates ranging from 9% to 27%. Infection can recur as long as four to 12 months after cessation of antibiotic therapy, so periodic follow-up and re-evaluation of ESR is essential to proper management of this disease.
Perichondritis and Chondritis
Bacterial infection of the perichondrium or cartilage is usually a result of accidental or surgical trauma to the auricle. In cases where it appears spontaneously, a high index of suspicion for overt or latent diabetes mellitus should be raised. The presenting complaint is pain and severe itching deep in the canal. The skin over the pinna is tender, indurated and edematous. In more advanced cases, the affected area may become crusted and weep purulent exudates with involvement of the surrounding soft tissues of the face and neck. Mild cases can be treated with debridement and topical and oral antibiotics. Antibiotics should be directed toward the offending organism from culture sensitivities. If the infection spreads to involve regional soft tissues and lymphatics, hospitalization and parenteral antibiotics are necessary. If subacute or chronic infection becomes established in the perichondrium or cartilage and continues despite treatment, surgical intervention under a controlled setting is indicated. Surgery involves excision of necrotic tissue with coverage by local skin flaps. Small irrigation drains should be placed beneath the flaps and irrigated with an antibiotic solution three times per day. The drains can be advanced as the condition resolves.
Relapsing polychondritis is an episodic and generally progressive inflammation of the cartilaginous structures of the body. An autoimmune etiology is suspected since many of these patients have circulating antibodies to type II collagen. Cartilage of the external ear, larynx, trachea, bronchi, and nose may be involved. During exacerbations, findings include fever, erythema, swelling, pain, anemia and an elevated ESR. With progression of disease, involvement of the larynx and trachea causes increased respiratory obstruction. Treatment is with oral corticosteroids.
Herpes Zoster Oticus (Ramsay Hunt syndrome)
Herpes zoster oticus is a viral infection of the ear caused by the varicella zoster virus. The virus causes infection along the dermatomes of one or more cranial nerves, thus describing shingles. In 1907, J.Ramsay Hunt, a neurologist described the cutaneous areas that can be involved in this infection. He noted that the trigeminal, geniculate, and upper cervical root ganglia might be involved, individually or combined. Ramsay Hunt syndrome has evolved to describe herpes zoster of the pinna with otalgia and facial paralysis. The earliest symptom is a burning pain in one ear, which may be accompanied by headache, malaise and fever for a couple of days. Vesicles usually appear 3 to 7 days after the onset of pain, and usually erupt on the antihelix, conchal bowl, and posterior lateral EAC. Infection of the geniculate ganglion may also present with facial paresis, or complete paralysis. In the case of complete paralysis, corneal protection with ophthalmic drops and lubrication at night is indicated. Oral steroids are also commonly prescribed and tapered over 10 to 14 days. Treatment with acyclovir, famcyclovir and valacyclovir have been shown to be effective in shortening the phase of viral shedding and reducing otalgia.
Erysipelas is an acute, localized but spreading superficial cellulitis that may involve the auricle. It is caused by group A beta hemolytic streptococci and characterized by involvement of the lymphatics. The affected skin is bright red, well demarcated and tender, with a distinctly advancing margin. Treatment with oral antibiotics should be started promptly, however, hospitalization and intravenous antibiotics may be necessary, if a rapid response is not evoked.
FACIAL NERVE TRAUMA
FACIAL NERVE PARALYSIS: OCULAR MANAGEMENT
The facial nerve is a very complex and unique nerve in both its anatomical course and function. Due to this complexity it is understandable that it is involved in many of the pathologic entities that affect the head and neck. One of these entities is trauma.
We will begin our discussion of facial nerve trauma by first examining the anatomy and function of the facial nerve. The anatomy of the facial nerve can be broken down into three major segments; intracranial, intratemporal, and extratemporal. The intracranial segment refers to that portion of the nerve that runs from the brainstem to the internal auditory canal (IAC). It can be further divided into two components, the motor root and the nervus intermedius. The motor root, as the name implies, carries the motor fibers of the facial nerve. The nervus intermedius on the other hand carries the facial nerve’s preganglionic parasympathetic fibers and special afferent sensory fibers. These two components join near the IAC to form the common facial nerve. The intratemporal segments begin as the nerve enters the IAC.
The first intratemporal segment is referred to as the meatal segment. It is the portion of the facial nerve traveling from the porus acusticus to the meatal foramen of IAC. It travels in the anterior superior portion of the IAC along with three other nerves; the superior vestibular nerve in the posterior superior portion, the inferior vestibular nerve in the posterior inferior portion, and the cochlear nerve in the anterior inferior portion. The length of the meatal segment is roughly 8-10mm. At the end of the IAC near the meatal foramen, the diameter narrows from 1.2mm to 0.68mm. This is the narrowest portion of the IAC and just so happens to be where the next segment of the facial nerve is located. This segment is the labyrinthine segment. It runs from the fundus to the geniculate ganglion, and is the shortest of all the intratemporal segments at 2-4mm in length. The geniculate ganglion houses the sensory and taste cells to the anterior 2/3 of the tongue and palate. It is also where the first branch of the facial nerve comes off of, the greater superficial petrosal nerve. This branch joins the deep petrosal nerve to form the vidian nerve, and is responsible for providing parasympathetic fibers to the lacrimal gland. The portion of the facial nerve that runs from the geniculate ganglion to the second genu is termed the tympanic segment. It is roughly 11mm in length, and is the most commonly injured portion of the facial nerve during middle ear/mastoid surgery. One of the major reasons for injury is secondary to the fact that the nerve is dehiscent in this area in 40-50% of the population. The next segment is termed the mastoid segment. It is the part of the facial nerve that runs from the second genu to the stylomastoid foramen, a length of roughly 12-14mm. Its course takes it between the incus and horizontal semicircular canal. It is at the end of this segment that the facial nerve gives off a branch to the stapedius muscle and the chorda tympani.
Once the facial nerve exits the stylomastoid foramen it gives off the postauricular nerve that supplies the external auricular and occipitofrontalis muscles as well as the branches to the posterior belly of the digastric and stylohyoid muscles. It then enters the parotid gland splitting the gland into a superficial and deep lobe. Within the parotid, the nerve splits into two major segments at a point termed the pes anserinus. The upper segment is termed the temporozygomatic segment, and the lower segment is termed the cervicofacial segment. These branches further split into the five major branches that supply the muscles of facial expression; the temporal, zygomatic, buccal, marginal mandibular, cervical branches.
The facial nerve fiber itself can be furthered divided anatomically. The three major components of the nerve fiber are the endonerium, perinerium, and epinerium. The endonerium surrounds each nerve fiber, and provide the endoneural tube. This tube needed for nerve regeneration. As such, if the endonerium is disrupted, the prognosis for return of function is worse. The perinerium surrounds a group of nerve fibers. It provides tensile strength, protects the nerve from infection, and provides pressure regulation. The last component is the epinerium. This is the layer that surrounds the entire nerve. It is responsible for providing nutrition to the nerve through the vasa nervorum.
The function of the facial nerve can be broken into three major areas, motor, sensory, and parasympathetic. The motor component supplies function to the muscles of facial expression as well as the stylohyoid, posterior belly of the digastric, stapedius and buccinator muscles. The sensory component can be divided into the special visceral afferent and general sensory afferent. The special visceral afferent is responsible for providing taste to the anterior 2/3 of the tongue. The general sensory afferent provides sensation to part of the tympanic membrane, the wall of the EAC, postauricular skin, and concha. The parasympathetic component provides secretory function to the submandibular, sublingual, and lacrimal glands as well as many of the seromucinous glands of the nasal and oral cavities.
Before we go into the details of facial nerve injury we must understand the basic classifications of nerve injury. A commonly used classification is the Sunderland Nerve Injury Classification. It is broken into five different classes. Class I injury is referred to as neuropraxia. It is a conduction block caused by the cessation of axoplasmic flow due to compression. This class of injury is typically what is felt when one’s leg “falls asleep.” A full recovery is expected with this injury. Class II is termed axonotmesis. In this injury axons are disrupted and Wallerian degeneration occurs distal to the site of the injury. The endoneural tube remains intact, so regeneration occurs. However, regeneration is very slow occurring at 1mm/day. One also expects complete recovery of function following class II injury. A class III injury is termed neurotmesis. In this injury, the neural tube is disrupted, thus regeneration potential and functional return are affected. If regeneration does occur, a high incidence of synkinesis exists. Synkinesis is the abnormal mass movement of muscles which do not normally contract together.
A class IV injury is classified as disruption of the perineurium, endoneurium, and axon. The epineurium remains intact. Poor functional outcome is expected if regeneration does occur with a high risk of synkinesis. Class V injury is the worst of the injuries, and is classified as complete disruption of the nerve. There is little chance of regeneration with this type of injury. The risk of painful neuroma formation is increased due to axonal sprouts that make their way out of the nerve sheath.
Trauma to the facial nerve is the second most common cause of facial nerve paralysis representing 15% of all cases of facial nerve paralysis. The most common site of injury in trauma is the temporal bone. There are many different types of trauma that lead to facial nerve paralysis. The ones that we will cover are temporal bone fractures, penetrating trauma, and iatrogenic trauma.
Most temporal bone fractures are due to blunt trauma and can be seen in up to 5% of all trauma victims. They are the most common traumatic cause of facial nerve paralysis. There are two distinct types of temporal bone fractures, longitudinal and transverse. Longitudinal fractures are the most common type making up 70-80% of all temporal bone fractures. The type of fracture seen is one that is parallel to the long axis of the petrous pyramid and results from blunt force delivered to the temporoparietal area. Typically, one can expect to see facial nerve paralysis in 25% of cases of longitudinal fractures. Transverse fractures are less common representing roughly 10-20% of all temporal bone fractures. The type of fracture seen is one that is perpendicular to the long axis of the petrous pyramid, and results from a frontal or occipital blow. This type of fracture results in facial nerve paralysis in 50% of cases. One may also see a mix of the two fracture types. This occurs in 10% of all temporal bone fractures.
In one study by Chang and Cass (1999), they reviewed the facial nerve pathologic findings of 67 longitudinal temporal bone fractures and 11 transverse temporal bone fractures where the patient was known to have facial nerve paralysis. In longitudinal fractures, 76% of cases showed bony impingement or intraneural hematoma while 15% showed a transected nerve. 9% either had no pathologic findings or just neural edema. In transverse fractures, 92% of cases showed transection of the nerve while 8% showed bony impingement or hematoma.
The next type of trauma that can result in facial nerve paralysis is penetrating trauma. This type of trauma typically affects the extratemporal segments of the facial nerve. However, gunshot wounds will cause both intratemporal and extratemporal injuries. Gunshot wounds to the temporal bone result in facial nerve paralysis in 50% of cases. This type of injury usually results in a much worse outcome than other types of trauma secondary to the fact that gunshot wounds typically result in a mixture of avulsion and blunt trauma to different portions of the nerve at the same time.
The next type of trauma is iatrogenic trauma. This can be further broken down into injury during surgery and birth trauma. Due to its complex course, the facial nerve is commonly encountered in many head and neck surgical procedures. The most common overall surgery where facial nerve injury occurs is the parotidectomy. The most common otologic procedure resulting in facial nerve injury is the mastoidectomy. Tympanoplasty and exostoses removal both account for 14% of cases of injury each. The mechanism of injury is either direct mechanical injury or heat generated from drilling near the facial nerve. The most common nerve segment injured during otologic surgery is the tympanic portion due to its high incidence of dehiscence in this area, and relation to surgical field. Nearly 80% of all cases of surgical related facial nerve injury go unrecognized. Birth trauma is another type of iatrogenic injury to the facial nerve. It is typically the result of a forceps delivery with compression of the facial nerve against the spine.
The work-up of facial nerve injury related to trauma begins with a good history and physical examination. Important aspects of the history include the mechanism (recent surgery, facial/head trauma), timing of injury (progressive loss of function or sudden loss), and associated symptoms (hearing loss or vertigo hint more toward a temporal bone injury). The physical examination must include a full head and neck examination looking for facial asymmetry and signs of facial injury (lacerations, hematomas, and ecchymosis). One must examine the head/scalp for signs of injury to help determine the vector of force if head trauma is involved. Otoscopic examination is another important aspect of the examination. Canal lacerations or step-offs as well as hemotympanum, tympanic membrane perforation, drainage of blood or clear fluid from middle ear may all be seen in temporal bone injury. Tuning fork tests (Weber/Rinne) with a 512 Hz fork can help determine if there is a conductive hearing loss. In addition, the muscles of facial expression should be closely examined. Dysfunction can be classified by the House-Brackmann Grading System. This system is divided into gross inspection and motion ability. During examination one must be aware that movement of the upper eyelid should not be considered a criteria for partial function since the levator palpebrae muscle helps in this function, but is innervated by CN III. The HB grading system can be found in the table below.
|I. Normal||Normal facial function in all areas|
|II. Mild dysfunction||Gross • Slight weakness noticeable on close inspection • May have slight synkinesis Motion • Forehead – Moderate-to-good function • Eye – Complete closure with minimal effort • Mouth – Slight asymmetry|
|III. Moderate dysfunction -First time you can notice a difference at rest||Gross • Obvious but not disfiguring difference between the two sides • Noticeable but not severe synkinesis, contracture, or hemifacial spasm • At rest, normal symmetry and tone Motion • Forehead – Slight-to-moderate movement• Eye – Complete closure with maximum effort • Mouth – Slightly weak with maximum effort|
|IV. Moderately severe dysfunction -First time you have incomplete eye closure -No forehead movement||Gross • Obvious weakness and/or disfiguring asymmetry Motion • Forehead – No motion • Eye – Incomplete closure • Mouth – Asymmetric with maximum effort|
|V. Severe dysfunction||Gross • Only barely perceptible motion • At rest, asymmetry Motion • Forehead – None • Eye – Incomplete closure • Mouth – Slight movement|
|VI. Total paralysis||No movement|
The work-up for traumatic facial nerve injury may also include radiographic evaluation. This typically involves CT and MRI scans. CT scans tend to be better for bony evaluation while MRI scans are utilized more for soft tissue detail and CPA pathology.
Another integral part of the evaluation of facial nerve injury is facial nerve testing. Testing has many functions. It is used to assess the degree of electrical dysfunction, helps with pinpointing the site of injury, and helps with determining treatment options. It can also be used to predict recovery of facial nerve function; partial paralysis is a much better prognosis tan total paralysis. The testing can be divided into two categories, topographic and electrodiagnostic tests.
Topographic tests are used to assess the integrity of specific facial nerve branches by testing the function of each branch. These tests are not utilized much anymore because they are not anatomically accurate and do not predict potential recovery of function.
The first of these tests is the Schirmer’s test. It is used to assess the function of the greater superficial petrosal nerve which is an evaluation of the protective mechanism of the eye. A piece of filter paper is placed in the conjunctival fornix of both eyes and the patient is asked to close his eyes for 5 minutes. After 5 minutes, the paper is removed from both eyes and the length of the areas that are moist is compared. An abnormal test can be defined as either a unilateral length measuring only 25% or less of the total length measured from both eyes or a total length from both eyes only measuring 25mm.
The next topographic test is the salivary flow test. It is used to test the integrity of the chorda tympani nerve by measuring the function of salivation with gustatory stimulation. The test begins with the cannulization of Wharton’s ducts. A gustatory stimulation is then applied. Measurements of salivary flow then occur over a 5 minute period. An abnormal test is defined as a reduction of 25% of the measured saliva when compared to the uninvolved side. The test is not utilized anymore since it is considered difficult to perform, causes significant patient discomfort and carries poor accuracy.
The third topographic test is the electrogustometry test. This test consists of stimulating the tongue electrically in order to produce a metallic taste. Both sides of the tongue are compared for results.
The next topographic test is the stapedial reflex or acoustic reflex test. This test is based on the ability of the stapedius muscle to contract in response to a loud sound. When contraction of the muscle occurs, the impedance of the middle ear changes. A loud sound is applied to one side, and the impedance of the middle ear from both sides is measured. If the stapedius muscle is out, there will be no impedance change on the affected side.
Electrodiagnostic tests utilize electrical stimulation to assess facial nerve function. All but the electromyography (EMG) test require a normal contralateral facial nerve to compare functional results with. The nerve excitability test (NET) compares the current thresholds required to illicit minimal muscle contraction on the normal side of the face to those of the paralyzed side. A stimulating electrode is applied over the stylomastoid foramen, and a DC current is applied percutaneously. The face is then monitored for movement. The electrode is then repositioned to the opposite side, and the test is performed again. A difference of 3.5 mA or greater between the two sides is considered significant. The main drawback to this test is that it relies on a visual end point making it very subjective.
The next test is the maximum stimulation test (MST). This test is similar to the NET, except that it utilizes maximal stimulation rather than minimal, and the main trunk as well as each major portion of the distal branches of the nerve are stimulated. The paralyzed side is compared to the contralateral side and the functional comparison is rated as equal, slightly decreased, markedly decreased, or absent. An equal or slightly decreased response is considered favorable for complete recovery. Markedly decreased or absent responses denote advanced degeneration with a poor prognosis. The response to this test becomes abnormal sooner than the response to the NET and is therefore considered superior to the NET. However, like the NET, this test is also subjective.
Electroneurography (ENoG) is another electrodiagnostic test utilized to assess the facial nerve. It provides quantitative analysis of the extent of degeneration without being dependent on observer qualification, and is thought to be the most accurate of the electrodiagnostic tests. The facial nerve is stimulated with an impulse applied at the stylomastoid foramen using bipolar electrodes. The summation potential is then recorded by a device utilizing bipolar electrodes placed near the nasolabial groove. The peak to peak amplitude of the evoked compound action potential is considered proportional to the number of intact axons. The two sides are then compared with the response on the paralyzed side of the face expressed as a percentage of the response on the normal side of the face. It is believed that surgical decompression of the nerve should be performed when 90% degeneration has occurred. Those with less than 90% degeneration within 3 weeks of facial nerve injury typically have an expected spontaneous rate of recovery of 80 – 100%. The disadvantages of this test include patient discomfort, cost, and test-retest variability that is due to positioning of the electrodes and excitation of the muscles of mastication.
The next test we will examine is electromyography (EMG). This test determines the activity of the muscle itself. A needle electrode is inserted into the muscle, and recordings are made during rest and voluntary contraction. Normally, voluntary movement will produce biphasic or triphasic potentials. When a lower motor neuron injury occurs, the muscles supplied by this nerve will undergo spontaneous movements called fibrillations that can be measured anywhere from 10-21 days following the injury. This test is typically not the first test utilized due to the amount of time needed to see signs of injury. Six to twelve weeks prior to the clinical return of facial function, polyphasic reinnervation potentials can be measured. These potentials are considered the earliest evidence of nerve recovery.
The treatment options for facial nerve injury differ by mechanism of injury and initial presentation. For instance, if the nerve is transected during surgery, it is recommended that the surgeon explore 5-10mm of the involved segment and stimulate both the proximal and distal segments. If there is a response with 0.05mA, full recovery and good function are expected, and thus further exploration is not required. However, if the nerve only responds distally a poorer prognosis is expected. As such, further exposure is warranted. If the loss of function is noted following surgery, wait 2-3 hours and then re-evaluate the patient. This should be ample time for any anesthetic to wear off. If the paralysis is still present following that time, the surgeon’s next move is based on the understanding of the integrity of the facial nerve. If the surgeon is unsure of the nerve’s integrity or the nerve was never identified during surgery, re-exploration is warranted. If the integrity of nerve is known to be intact, treatment can begin with a high dose of steroids. This is typically prednisone at 1mg/kg/day for 10 days and then a taper. After 72 hours, EnoG is utilized to assess the degree of degeneration. If there is greater than 90% degeneration, one should re-explore. If there is less than 90% degeneration, one can just monitor the patient. If worsening paralysis occurs the surgeon should re-explore. However, if no regeneration occurs, but the function does not worsen, the timing of exploration or whether to explore is controversial.
Quaranta et al (2001) examined the results of 9 patients undergoing late nerve decompression (27-90 days post injury) who all had greater than 90% degeneration. Seven of the patients achieved HB grade 1-2 after 1 year, and the other 2 patients achieved HB grade 3. They concluded that patients may still have a benefit of decompression up to 3 months out. Shapira et al (2006) performed a retrospective review looking at 33 patients who underwent nerve decompression. They found no significant difference in overall results between those undergoing early (<30 days post-injury) vs. late (>30 days post-injury) decompression. Most studies like these have been very small and lack control groups. Some studies have shown improvements with decompression occurring 6-12 months post-injury, but further evidence is required.
If there is facial nerve paralysis following birth or extratemporal blunt trauma, it is recommended that there be no surgical exploration since greater than 90% of these cases are expected to regain normal to near normal function. When dealing with temporal bone fractures the degree of paralysis guides the treatment options. If complete paralysis following a temporal bone fracture occurs then complete transection of the nerve must be assumed. For this reason exploration is warranted. If there is a partial or delayed loss of function following the fracture the treatment begins with high dose steroids and ENoG testing after 72 hours. If there is greater than 90% degeneration, explore. If there is less than 90% degeneration, one can monitor and explore at a later date depending on worsening or failure to regenerate.
When dealing with penetrating trauma there is a high likelihood of nerve transection, thus exploration is usually warranted. If the injury occurs in the extratemporal segments exploration is typically not recommended when the injury occurs distal to the lateral canthus since the nerve endings are very small and there is a rich anastomotic network from other branches in this area. However, when exploration is going to occur it should take place within 3 days of injury because the distal branches can still be stimulated, thus making it easier to locate them. With gunshot wounds, however, delayed exploration is actually recommended as this type of trauma results in extensive nerve damage, and waiting a little longer to indentify the extent of injury can be beneficial in forming a surgical plan.
If decompression of the nerve must occur, the patient’s auditory and vestibular function must be taken into account. If the auditory and vestibular function is intact, a transmastoid/middle cranial fossa approach is warranted. If the auditory and vestibular function is absent, a transmastoid/translabyrinthine approach is recommended. Since the nerve may be injured along multiple segments all attempts should be made to localize the injured site pre-operatively. This will make a big difference in the amount of exposure required while potentially decreasing the morbidity of the procedure. However, this may not be possible, and full exposure of the nerve from the IAC to the stylomastoid foramen may be required. During the decompression, diamond burs and copious amounts of irrigation should be utilized to prevent thermal injury. The thin layer of bone overlying the nerve is typically bluntly removed.
Whether to perform neurolysis or not to open the nerve sheath is debatable. However, if a hematoma is identified it should be drained.
If repair of the facial nerve is required, there are many options available to the surgeon. With neural repair, the surgeon should expect to start seeing some recovery starting around 4-6 months. The recovery can last up to 2 years following repair. The surgical options available differ based on the timing of injury since after 12-18 months, muscle reinnervation becomes less efficient even with good neural anastomosis.
When it comes to nerve repair, the goal is a tension free, healthy anastomosis. The rule is to repair earlier than later, but the exact timing of the repair is controversial. Some authors have reported improvement with repairs as far out as 18-36 months. May and Bienstock recommend repair within 30 days, but others have found superior results if done up to 12 months out. After 2 weeks of injury, collagen and scar tissue replace axons and myelin. For this reason unhealthy nerve endings must be excised prior to anastomosis.
Primary anastomosis should be attempted first since it provides the best overall results of any surgical intervention. It can be performed when the defect is less than 2cm since mobilization of the nerve can give nearly 2cm of length. However, if more than this is mobilized, the risk to further neural injury increases secondary to devascularization. The most important aspect of neural repair is ensuring that the endoneurial segments are aligned as this will promote regeneration. The nerve ends should be sutured together using three to four 9-0 or 10-0 monofilament sutures to bring the epineurium or perineurium together.
If the defect is greater than 2 cm or a tension free anastomosis cannot be obtained, then nerve grafting or transfer should be performed. The problem with this is that this results in partial or complete loss of the donor nerve function. The type of grafting/transfer performed depends on whether both the proximal and distal segments of the nerve are available. If they are available, then a simple graft can be used to bring them back together. A commonly used nerve for grafting is the great auricular nerve since it is usually in the surgical field already. It is located within an incision made from the mastoid tip to the angle of the mandible. However, only 7-10cm of this nerve can be harvested. The complication from this harvest is a loss of sensation to lower auricle. Another commonly used nerve for grafting is the sural nerve. It is located 1 cm posterior to the lateral malleolus, and can provide 35cm of length making it extremely useful in cross facial anastomosis. The complication that occurs with its use is the loss of sensation to lateral calf and foot. One study quoted that 92-95% of patients undergoing graft repair when proximal and distal portions of the nerve are available have some return of facial function. Of those, 72-75% have good results (HB 3 or above).
If the distal nerve segment is the only segment available, the surgeon must ensure that the facial musculature is suitable for reinnervation. This is done through EMG testing and/or muscle biopsy. The options for repair in this situation typically involve the use of the hypoglossal nerve or the contralateral facial nerve. The direct hypoglossal-to-facial graft is performed by attaching the distal segment of the injured facial nerve directly to the hypoglossal nerve. With this type of graft, 42-65% of patients are expected to experience decent symmetry and tone. However, the complications are difficult to deal with and include atrophy of the ipsilateral tongue and difficulties with chewing, speaking, and swallowing. A more tolerated grafting technique is the partial hypoglossal-to-facial jump graft. This is done by the use of a nerve cable graft (usually the sural nerve) to connect the distal end of the facial nerve to a notch in the hypoglossal nerve. It results in much fewer complications, but increases the recovery time. May compared results of the direct VII-XII graft to the VII-XII jump graft. In his study, only 8% of patients experienced permanent complications from loss of the hypoglossal nerve in the jump graft compared to 100% in the direct graft. Of those patient undergoing jump grafting, 41% obtained good movement with less synkinesis. The motor function, however, was not as strong in the jump graft group.
Another option is the facial-to-facial graft. This procedure can either involve a single contralateral branch connected to the distal nerve or multiple anastomoses from segmental branches to segmental branches. The best results from this type of procedure have been seen when a sural nerve graft is utilized to connect the buccal branch on the contralateral side to the distal nerve stump. The VII-VII graft technique is typically not recommended secondary to the weakness caused to the contralateral facial nerve and lack of power to control the musculature resulting in poor results.
The next topic to be covered is one that many head and neck surgeons have found very useful when dealing with the facial nerve; facial nerve monitoring. The first monitors that were utilized relied on sensing muscle movement. They are rarely used now since a large threshold must be reached to illicit movement, and by that time injury may have already occurred. Also, there is a poorer response to facial nerve stimulation than what is seen in electrophysiologic techniques. The principle technique utilized now days involves electromyography. Electrodes are used to detect differences in electrical potential associated with a depolarizing current. A graphic and acoustic signal are then recorded.
There are two types of responses that one can expect. The first are repetitive responses. They occur following the cessation of surgical manipulation, and represent irritability of the nerve secondary to nerve injury. They are used to warn the surgeon of injury or impending injury. The next type of responses are the nonrepetitive responses. They are single responses secondary to direct mechanical or electrical stimulation, and are utilized to map the course of the nerve.
There are many uses for facial never monitors. These include identity, mapping, injury identification, and prognosis. The monitors can be utilized to help identify exactly where the nerve is located. This can be done through mechanical or electrical stimulation. Once located, the nerve can then be mapped by repeated stimulation. Most surgeons advocate bipolar stimulation as it is more precise, but does carry more false-negatives than monopolar techniques. Injury identification relies mainly on repetitive responses as described above. This allows the surgeon to alter his or her actions to prevent or lessen injury. The monitors can also provide prognostic information of facial nerve function following surgery. There are two different measurements that can be taken to determine prognosis. The first is a technique that utilizes stimulated compound action potentials. It is the least utilized as it relies on proper electrode placement, thus leading to poor reproducibility. During this technique, a 0.4mA stimulus is applied to the nerve and a compound action potential is recorded. If the compound action potential is greater than 500-800 microvolts, a HB I-II is expected. However, as the action potential drops below 500 microvolts, the outcome becomes poorer. The most commonly used technique for determining facial nerve prognosis with monitors is termed the nerve stimulus threshold. This technique utilizes an electrical stimulus applied to the proximal end of the nerve. If the nerve responds with a stimulus that is less than 0.3mA, a HB I-II is expected. If a greater then 0.3mA stimulus is required to stimulate the nerve, one can expect a HB III-V.
Many have wondered whether the use of a facial nerve monitoring during surgery really makes a difference to the overall outcome following surgery. Dickinson and Graham reviewed the use of facial nerve monitors in surgical cases involving the excision of CPA tumors. Thirty-eight cases were performed without facial nerve monitoring, 29 cases with a pressure or strain gauge sensor, and 41 cases with monitoring by EMG. They reported poor outcomes (HB V-VI) in 37% of cases where no monitor was used and 21% of cases where the older pressure or strain gauge sensor was utilized. However, poor outcome was only noted in 4% of cases where EMG was utilized. A confounder with this study was that there was a higher incidence of larger tumors in the unmonitored group.
Pensak et al examined 250 cases involving surgery on chronic middle ear disease in which all cases were monitored. They reported that in 100% of cases the facial nerve was grossly identified. However, only 82% were confirmed with monitor stimulation. In cases where the nerve was exposed, they reported that the monitor alerted the surgeon to its location in 93% of cases. Silverstein and Rosenberg examined 500 cases in which facial nerve monitoring was utilized. They reported no cases of facial nerve injury, but did report that the monitor prevented injury to the facial nerve in 20 cases.
Terrell et al examined 117 cases of parotid surgery where 56 cases had monitors and 61 cases were performed without a monitor. They reported a statistically significant decrease in the rate of post-operative paresis in the monitored group, but found no difference in long term outcome. They did find that longer OR times were associated with decreased rates of post-operative paresis. Witt reviewed 53 cases of parotid surgery in which 33 had monitors and 20 did not. He found no difference in paresis rates as well as no difference in long term outcome.
Another question that is commonly asked when it comes to facial nerve monitoring is the safety of its use. Most ask the question “does repetitive stimulation lead to facial nerve injury?” Babin et al examined the use of pulsed current stimulation to stimulate cat facial nerves. A pulse of 1mA was applied to the nerve every 3 seconds for 1 hour. A transient decrease in nerve sensitivity following cessation of the stimulus was noted, but no permanent injury was reported. Hughes et al examined the use of pulsed and constant current models for stimulation of mouse sciatic nerves. In all cases in which pulsed current was utilized, no injury was reported. In some cases in which constant current was utilized, mild injury and axonal degeneration occurred. However, nearly all monitors now utilize pulsed currents. For this reason, the surgeon must be cognizant of the type of current being applied for stimulation.
In conclusion, the facial nerve is very complex in its function and anatomical course. As such, it can be easily injured by many mechanisms. A good understanding of its function and anatomical course can make it easier for the surgeon to prevent injury and indentify where an injury has occurred. Combining this knowledge with the knowledge gained from diagnostic testing, a surgeon can come up with a good treatment plan.
FACIAL NERVE PARALYSIS: OCULAR MANAGEMENT
Ocular complications from facial nerve paralysis can be quite devastating. Facial nerve paralysis results in cosmetic as well as functional problems. With facial nerve paralysis the upper lid fails to drop down and close. The lower lid may become lax and evert, developing a lower lid ectropion. Paralysis of the upper lid leads to lagopthlamos, which results in incomplete closure of the lid over the cornea. This can cause the cornea to dry, disrupting the tear film and results in pain, corneal ulceration, infection and even corneal perforation. Pre-injury factors that lead to increased risk of complications are the lack of a good Bell phenomenon, corneal anesthesia and pre-injury dry eye.
Normal eye closure is facilitated by the action of the obicularis oculi. Contraction of this muscle results in lowering of the upper lid. The lower eye lid has a minimal contribution to eye closure. The eyelid functions to protect the eye from light, foreign material, injury and desiccation. It also functions in the distribution of tears and maintenance of the tear film. The skin of the eyelids is very thin, with the upper lid being thicker than the lower lid. There is little subcutaneous fat and the skin is adherent to the tarsal plate.
The normal horizontal distance of the eye is 28-30mm. The distance from the upper lid margin to the supratarsal crease is 10mm. The margin reflex distance, or the distance from the corneal reflex to the lid margin, is approximately 4-5mm for the upper lid and 5mm for the lower lid. The lid margin of the lower lid rests at the limbus of the eye. The upper lid margin rests approximately 2mm below the limbus.
The tarsus, or tarsal plate, is a dense fibrous tissue that provides contour and a fibrous skeleton to the lids. The tarsus is 25mm in length and 1mm thick. The tarsus contains the Meibomian glands. The height of the upper tarsus is 10mm and the lower tarsus is about 4mm.
The main protractor of the eye is the obicularis and is innervated by the facial nerve. The main retractors of the eye are the levator and Mullers muscle, innervated by the oculomotor nerve. The obicularis has three components: the orbital part, the pretarsal and the preseptal parts. The pretarsal and preseptal components are responsible the eye closure. The preseptal and pretarsal components extend medially to form the medial canthal tendon. The same areas extend laterally to form the lateral canthal tendon. The lateral canthal tendon attached to Whitnalls tubercle on the medial aspect of the lateral orbital wall. The lateral canthus attachment is slightly higher than the medial canthal attachment. The medial canthal complex is important for normal lacrimal functioning. The lacrimal sac is situated between the anterior limb and posterior limb of the medial canthal tendon. When blinking occurs, the tears are forced to the lacrimal sac through the canaliculus. When the eye opens, the sac is compressed and the tears are forced out of the sac to the inferior meatus.
When eye paralysis occurs with impaired facial nerve function treatment should begin immediately. Initial treatment consists of using ophthalmic drops and ointments, taping the eye, soft contact lenses, scleral shields or moisture chambers. Also of utility is the tarsorrhaphy suture. Despite the initial management of the eye, the majority of patients will require definitive surgical treatment to prevent permanent damage to the cornea. Surgical treatments include the temporalis muscle transfer, encircling the eye with fascia lata or silicone, palpebral springs, tarsorrhaphy, lower lid shortening, lid loading or combinations of these procedures.
Palpebral springs are placed under the skin of the upper lid. They act to force the eye closed with relaxation of the levator and Mullers muscle. Palpebral springs are less visible, but are technically more difficult to place and have a higher rate of extrusion.
Tarsorrhaphy works well for lower lid ectropion. However, it is cosmetically less appealing and often results in loss of peripheral vision.
Lower lid shortening procedures consist of wedge excision with lateral canthopexy. This procedure works well for lower lid ectropion and as an adjunct to upper lid loading.
The early technique of upper lid loading consisted of making an incision in the pretarsal crease and forming a pocket in the subcutaneous tissue. A weight was then placed in the pocket and the skin closed. This technique resulted in frequent mobility of the weight and extrusion. The earlier weights were stainless steel. These weights had a higher profile, were migratory and had a high rate of extrusion. Later, the gold weight was used. It has a higher density so a heavier weight of the same size can be used. It is also malleable, so it can be conformed to the contour of the globe. It was also mobile and a high rate of extrusion.
The later technique of upper lid loading consist of making an incision in the upper pretarsal crease and incising through the levator aponeurosis. The tarsal plate is exposed and the weight is placed directly on the tarsal plate and fixed with sutures. The levator aponeurosis, muscle and skin are then closed in layers. This provides better coverage of the weight and prevents extrusion and migration and the plate is less visible. The technique is straight forward and consistent. Though gold remains the implant material of choice, recently a titanium chain has been developed.
Kinney et al described an algorithm for the management of lid paralysis. Surgical techniques described included auricular cartilage grafting, tarsorrhaphy, lateral canthotomy, elevation of the subobicularis oculi fat and brow lifting. Ultimately, gold weight implantation was necessary.
Snyder et al investigated the timing of gold weight implantation. They evaluated the outcomes and complications of early (<30 days) and late (>30 days) gold weight implantation. They found that 89.2% achieved satisfactory eye closure. The early and late groups were statistically similar in eye closure and complication rates.
Foda investigated upper lid gold weight implantation and lower lid shortening. He found that this combination resulted in complete correction of lagophthalmos and ectropion. Pre op symptoms resolved in 92.5% of the patients studied.
Harrisberg reported a series of 103 patients with upper lid gold weight implants. Forty six patients had their weights removed. Of these, 78% were removed because of facial nerve recovery. The remaining 22% were removed because of cosmetic dissatisfaction, implant becoming too superficial, migration and partial extrusion. The majority of these implants were placed using the earlier method.
In a series 16 patients, Chepeha reported improvement in lagophthalmos, corneal coverage and a high patient satisfaction. The lagophthalmos improved from an average of 7.5mm to 0.5mm. Corneal coverage improved from 73% to 100% coverage. There were no extrusions in this series.
In conclusion, upper lid loading with gold weights is safe and effective. They may be implanted early and are reversible. When used in combination with lower lid shortening, upper lid gold weights provide excellent cosmetic and functional results in the paralyzed eye.
GENETIC HEARING LOSS
Hearing loss is the most common sensory deficit in humans. Roughly one child in a thousand is born with hearing impairment significant enough to compromise the development of normal language skills. Hearing loss can be caused by environmental factors as well as genetic factors. Environmental causes include pre- and post natal infection and ototoxic drug exposure. But it is estimated that 50% to 75% of all childhood deafness is due to hereditary causes. There are two main forms of genetic hearing loss, syndromic and nonsyndromic. Children with syndromic hearing loss have other clinical features in addition to the hearing loss. About 15-30% of the hereditary hearing loss is syndromic, while the vast majority is nonsyndromic (70%).
Hereditary hearing loss can be transmitted in several inheritance patterns, including autosomal dominant, autosomal recessive, X-linked inheritance and mitochondrial inheritance. Briefly, autosomal dominant inheritance exhibits a vertical pattern of transmission. Only one altered gene is needed for an individual to be affected. The offspring has 50% chance to receive the copy of the mutated gene from the affected parent. The most common pattern of transmission in hereditary hearing loss is autosomal recessive. A child must have both copies of the mutated gene to exhibit deafness. The parents will most likely have normal or near normal hearing even though they possess the recessive gene. Typically there is a 25% chance that the offspring will be affected and manifest hearing impairment or deafness. This mode of transmission has a horizontal pattern. The X-linked inheritance involves particular genes located on the X chromosome. It more commonly affects male because they possess a single X chromosome and will present phenotypically with any mutation on the X chromosome. Female can carry the mutation on one of the X chromosomes without phenotypic expression. Her sons have a 50% chance of inheriting the mutation and express phenotypically. The daughters have a 50% chance of inheriting the mutation and become a carrier of the mutation. Mitochondrial inheritance is caused by mutations in the mitochondrion DNA, small loop of DNA in mitochondrion. Only mothers can pass on the mutation because only the eggs carry mitochondrion DNA.
When two genetic loci lie near each other on a chromosome, they tend to be inherited together. Based on this principle, genetic linkage analysis is used to localize the disease gene to a specific region on a specific chromosome. There are also hundreds of mouse models available for studying genetic hearing loss. Each provides a piece of the puzzle in our understanding of inner ear biology. Some are directly relevant to human deafness and others provide key elements in the development and function of sensory structures of the ear.
Hereditary deafness is genetically a highly heterogeneous disease with many different genes responsible for auditory dysfunction. Genetic heterogeneity has been observed repeatedly in our improved understanding of syndromic hearing loss. Mutations of several different genes can cause the identical clinical phenotype. On the other hand, different mutations in one gene can cause variable phenotypes.
Syndromic hearing loss
There are over four hundreds syndromes with hearing loss have been described. The Online Mendelian Inheritance in Man (www.ncbi.nlm.nih.gov/Omim) has comprehensive descriptions of the clinical features and molecular genetics of these syndromes as well as an all-inclusive list of references.
It is a hereditary disorder of basement membranes. Majority of the cases are inherited in an X-linked manner (80%). Autosomal recessive as well as autosomal dominant inheritance has also been observed. It is characterized by the association of renal dysfunction with hearing loss. Symptoms include high frequency sensorineural hearing loss, hematuria with progressive renal failure, and ocular abnormalities. Microscopic hematuria was found to be the most reliable urinary criterion of hereditary nephritis in both males and females. The hematuria was often accompanied by red cell casts, indicating that the renal lesion is a glomerulitis. Men were more severely affected than women. They had striking urinary abnormalities in early childhood which progressed to renal failure in adulthood. Affected women had less obvious urinary findings and rarely developed uremia. Typical ocular associations are a dot-and-fleck retinopathy, which occurs in approximately 85% of affected adult males, anterior lenticonus, which occurs in approximately 25%, and the rare posterior polymorphous corneal dystrophy.
Mutations in three different type 4 collagen genes (COL4A3, COL4A4 and COL4A5) have been identified. These collagens are found in the basilar membrane, parts of the spiral ligament, and stria vascularis. Although the mechanism of hearing loss is not known, it is suggested that the loss of integrity of the basement membrane might affect adhesion of the tectorial membrane. Also in the glomerulus there is focal thinning and thickening with eventual basement membrane splitting.
It is an autosomal dominant disorder. The main clinical features include branchial derived anomalies, otologic anomalies, and renal malformation. The hearing loss found in BOR is highly variable. The loss can be sensorineural, conductive or mixed. It can be stable or progressive and the severity can range from mild to profound. Ear pits are found in about 80% of the BOR cases. Of those persons with ear pits, some have only one while others have one in front of each ear. Some individuals have deformed outer-ear(s), middle-ear(s), and/or inner-ear(s). About 60% of the BOR cases have branchial cysts or fistulas which are small holes located on the front, external lower third of the neck. People with branchial cysts or fistulas usually have two (one on each side); however, some only have one. Because they often become infected, branchial cysts or fistulas are often surgically removed. Kidney anomalies are found in about 15% of those with BOR. Most of the anomalies have minimal clinical significance and consists of minor changes in the anatomy of the kidney or urine collection system. Kidney function is normal in these individuals. More severe kidney anomalies have been reported. This range from small, normal functioning kidneys all the way to the kidneys (one or both) being absent. Three of the less common symptoms associated with BOR are ear tags, blocked tear ducts, and short palate.
Branchio-oto-renal syndrome is caused by mutations in EYA1 gene. Expression pattern of this gene indicated its role in development of the inner ear and kidney. EYA1 homozygous knockout mice lack ears and kidneys because of apoptotic regression of the organ primordial.
Jervell and Lange-Nielsen syndrome
It is an autosomal recessive disorder. Persons affected by this syndrome have prolongation of the QT interval, torsade de pointe arrhythmias (turning of the points, in reference to the apparent alternating positive and negative QRS complexes), sudden syncopal episodes, and severe-to-profound sensorineural hearing loss.
Two genes, KVLQT1 and KCNE1, have been identified. Little is known about the function of these genes. KVLQT1 is expressed in the stria vascularis of mouse inner ear. Within the inner ear, endolymph homeostasis is controlled in part by the delayed rectifier potassium channel. This channel is formed by proteins encoded by the KVLQT1 and KCNE1 genes. Heterozygotes of KVLQT1 have Romano-Ward syndrome, which unlike JLN syndrome, does not include severe-to-profound hearing loss in the phenotype.
It is X-linked inheritance. Classic features include specific ocular symptoms (pseudotumor of the retina, retinal hyperplasia, hypoplasia and necrosis of the inner layer of the retina, cataracts, and phthisis bulbi), progressive sensorineural hearing loss, and mental disturbance, although less than one-half of patients are hearing impaired or mentally retarded.
NDP gene has been identified in association with this disease. It encodes for norrin, a protein that has homologies at the C-terminus to a group of proteins including mucins. The exact mechanism of hearing loss remains to be determined. It has been suggested that norrin protein regulates vascularization of the cochlea and retina.
Pendred syndrome is the most common syndromal form of deafness, accounting for about 4-10% of cases. It is an autosomal recessive disorder and represents with goiter and sensorineural hearing loss. Usually the goiter is clinically evident at approximately 8 years of age, but adult onset has also been noted. Patients are usually euthyroid but can be hypothyroid. A positive potassium perchlorate discharge test may be helpful in identifying an organification defect, the defective organic binding of iodine in the thyroid gland. But the test is nonspecific and its sensitivity is unknown. Hearing loss is usually congenital, bilateral, severe to profound, and sloping in the higher frequencies. It may be fluctuating or progressive. Variable vestibular dysfunction has been exhibited by the patients. Enlargement of the vestibular aqueducts and the endolymphatic sac and duct are found in nearly all patients. An incomplete partition of the cochlear may be present and the vestibule may be enlarged.
Mutations of the gene PDS on chromosome 7q31 account for most cases of Pendred syndrome. Mutations in this gene can also be found in nonsyndromic deafness DFNB4 and many cases of enlarged vestibular aqueduct syndrome. The PDS gene encodes pendrin, an anion transporter found in the inner ear, thyroid, and kidney. It possibly plays a role in endolymphatic fluid resorption.
A PDS knockout mouse was generated in 2001 by Everett, which showed complete deafness. The inner ear of the mouse model was anatomically normal until embryonic day 15, at which time all the endolymph-containing spaces start to enlarge. Inner and outer hair cell degeneration also observed under electron microscopic studies. Interestingly, no thyroid abnormalities have been observed with this mouse model.
It is an autosomal dominant disease. Three phenotypes have been identified. Classic features of this syndrome include progressive myopia beginning in the first decade of life and resulting in retinal detachment and blindness, vitreoretinal degeneration, premature joint degeneration with abnormal epiphyseal development, midface hypoplasia, and irregularities of the vertebral bodies, cleft palate deformity and variable sensorineural hearing loss.
It is caused by mutations in COL2A1, COL11A2, or COL11A1. The classic phenotype is associated with mutations in COL2A1, a fibrillar collagen that is arrayed in quarter-staggered fashion to form fibers similar to those of COLI. Mutations in COL11A2 cause STL3, a disease characterized by the typical facial features of STL1 in combination with hearing impairment. Cleft palate and mild arthropathy also occur, however ocular signs are absent.
Treacher Collins syndrome
Treacher Collins syndrome is a disorder of craniofacial development. It is autosomal dominant inheritance with variable expression. The features include slanting of the eyes with inferior displacement of the lateral canthi with respect to the medial canthi, coloboma of the lower lids, micrognathia, microtia and other deformity of the ears, hypoplastic zygomatic arches, and macrostomia. Conductive hearing loss and cleft palate are often present.
TCOF1 gene was identified. It has suggested that TCOF1 plays a role in nucleolar-cytoplasmic transport. All of the mutations observed resulted in introduction of premature termination codons into the reading frame and cause premature termination of the protein product.
It is named for Charles Usher (1914), a British ophthalmologist who emphasized their hereditary nature. It is an autosomal recessive disorder. It is characterized y hearing loss and retinitis pigmentosa. Clinically three different types can be distinguished. They are distinguished based on the severity or progression of the hearing loss and extent of the vestibular system involvement. Type 1 patients have congenital profound congenital deafness, absent vestibular response and the onset of retinitis pigmentosa is in the first decade of life. Type 2 patients have normal vestibular response. Type 3 patients have progressive hearing loss, variable vestibular response and variable onset of retinitis pigmentosa.
A total of 11 loci and 6 genes have been identified in Usher syndrome. Among these, MYO7A, which belongs to the family of unconventional myosins, encode myosin 7A. It represents a unique molecular motor for hair cells. In the mutant mice, the organization of stereocilia bundles is disrupted, impairing the function of hair cells. CDH23 gene encodes cadherin 23, which is an adhesion molecule maybe important for crosslinking of stereocilia. It may also be involved in maintaining the ionic composition of the endolymph. Studies have suggested that myosin 7A, cadherin 23, along with another protein harmonin, which in coded by USH1c gene, form a transient functional complex in stereocilia.
The disease was named for Petrus Johannes Waardenburg, a Dutch ophthalmologist (1886-1979) who was the first to notice that people with two different colored eyes frequently had hearing problems. The clinical features usually include dystopia canthorum, meaning the lateral displacement of the inner canthus of the eyes to give an appearance of a widened nasal bridge, pigmentory abnormalities of the skin, iris, and hair, and sensorineural hearing loss. There is a great deal of variation in the hearing loss. Some of the affected persons escape deafness. There are four subtypes of Waardenburg syndrome have been described. Type 1 typically shows wide confluent eyebrow, high broad nasal root, heterochromia irides, brilliant blue eyes, premature gray of hair, eyelashes, or eyebrows, white forelock, and vestibular dysfunction. Type 2 is like type 1 but without dystopia canthorum. The penetrance of hearing loss is slightly higher in type 2 than type1. Type 3 is also called Klein-Waardenburg syndrome. It has type 1 clinical features and hypoplastic muscles and contractures of the upper limbs. Type 4 is also called Shal-Waardenburg syndrome. It has type 2 clinical features and Hirschsprung’s disease.
Five genes on five chromosomes have been identified. PAX3 gene mutations have been indicated to be associated with type 1 and type3 phenotypes. PAX3 is a DNA-binding protein that is important in determining the fate of neural crest cells in the developing nervous system. Type 2 has been linked to MITF gene mutation. Three genes have been associated wit type 4 phenotypes. They are EDN3, EDNRB, and SOX10. In vitro studies of EDN3 showed a stimulation of proliferation and melanogenesis of neural crest cells. EDNRB is postulated to play an essential role in the development of the two neural crest-derived cell lineages, epidermal melanocytes and enteric neurons. SOX10 belongs to the family of transcription factors that bind DNA and regulate its transcription. Interaction of these three genes has been suggested. The genetics of Waardenburg syndrome highlight the principle of genetic heterogeneity.
Nonsyndromic hearing loss
According to the Hereditary Hearing Loss homepage (www.uia.ac.be/dnalab/hhh/ ), 80 loci for nonsyndromic hearing loss have been mapped to the human genome. And 30 genes have been identified. Based on the type of gene product, these genes can be categorized into several groups. 1) channel and gap junction components, 2) myosin and other cytoskeletal proteins, 3) transcription factors, 4) extracellular matrix proteins, 5) unknown function genes.
There are autosomal dominant, autosomal recessive and X-linked forms of nonsyndromic hearing loss. In general, recessive inheritance shows prelingual onset of hearing loss. And the severity is severe to profound with all frequencies affected. In autosomal dominant forms, the phenotype is less severe. The onset is usually postlingual. The severity is ranging from moderate to severe. Hearing loss is seen in middle, high, or all frequencies with only three loci having hearing loss in the low frequency range.
Genes for homeostasis
Maintaining ion homeostasis within the cochlear duct, especially the high potassium concentration in the endolymph, is of extreme importance for signal transduction involved in the hearing process. Potassium recycling pathway is postulated to start with an efflux of potassium from the outer hair cells through potassium channel. Through gap junctions between the supporting cells, the ions migrate to the stria vascularis, for where they are secreted into the endolymph.
Four connexins have been implicated: connexin 26 (GJB2), connexin 31 (GJB3), connexin 30 (GJB6), and connexin 43 (GJA1). Connexins belong to a family of gap junction proteins responsible for the intercellular transport of ions, metabolites, and second messengers. GJB2 is the first nonsyndromic sensorineural deafness gene to be identified. It is estimated that mutations of GJB2 accounts for 50% of recessive nonsyndromic hearing loss. GJB2 encodes connexin 26. Animal studies to examine the role of connexins in the cochlea have been undertaken only for this protein. It is expressed in stria vascularis, basement membrane, limbus, and spiral prominence of cochlea. Rat immunohistochemical finding suggest that it plays a role in recycling of potassium ions back to the endolymph of the cochlear duct after stimulation of the sensory hair cells. 80 recessive and 6 dominant mutations have been found in GJB2. One mutation, 35delG (one guanosine residue deletion from nucleotide position 35), is very frequent in Caucasian population. This mutation results in shifting of the reading frame and protein truncation. Because of the high prevalence of this mutation in Western countries and the small size of the GJb2 gene, diagnostic testing is available.
KCNQ4 is a component of a potassium channel. It is though to be involved in the potassium recycling pathway as well. SLC26A4 encodes an anion transporter. This mutation gives rise to a broad spectrum of clinical symptoms: from nonsyndromic SNHL to hearing loss with enlarged vestibular aqueduct to full-blown Pendred syndrome with thyroid goiter.
Four transcription factor genes have been identified: POU3F4, POU4F3, EYA4, and TFCP2L3. POU3F4 is responsible for X-lined mixed hearing loss. The conductive hearing loss is from stapes fixation. These patients suffer from an increased perilymphatic pressure causing the typical “gusher” that appears during stapes footplate surgery. POU4F3 is associated with autosomal dominant hearing loss. Its knockout mice fail to develop hair cells with subsequent loss of spiral and vestibular ganglia. No gross morphological changes were seen in Pou3f4 knockout mice, but ultrastructural alterations wee found in the spiral ligament fibrocytes. Both EYA 4 and TFCP2L3 cause autosomal dominant hearing loss.
Maintaining the highly organized and specialized structures like actin-rich stereocilia of the sensory hair cells is very important for the processes of hearing and balance. Among the cytoskeletal components involved in nonsyndromic hearing loss are myosins, otoferlin, actin-polymerization protein, harmonin, and cadherin. Four unconventional myosins genes (MYO3A, MYO6, MYO7A, MYO15) and one conventional myosin gene MYH9 have been identified. Myosins are molecular motor proteins that bind to actin and that hydrolyze ATP to generate the force to move across actin filaments. Unconventional myosins move macromolecules along actin filaments. They are important for the structural integrity of the stereocilia. All the unconventional myosin mouse models display the vestibular dysfunction.
Otoferlin acts in calcium triggered synaptic vesicle trafficking. It is encoded by OTOF. Recently it has been reported to account for 4.4% of recessive nonsyndromic hearing loss negative for GJB 2 mutations in Spanish population. If this finding is confirmed in other population, it might be one of the genes with putative diagnostic implications.
Harmonin organizes multiprotein complexes in specific subcellular domains. They anchor and cluster transmembrane proteins and recruit signaling molecules. They are also associated with Usher type 1C disease.
Cadherins are components of adherens junctions and play critical roles during embryogenesis and organogenesis. It is encoded by CDH23. In mouse model, the mutations showed disruption of stereocilia organization during early hair cell differentiation. It is also associated with Usher type 1D disease.
Extracellular matrix components
The tectorial membrane overlying the hair cells plays a crucial role in the mechanosensory transduction process. TECTA gene encodes alpha tectorin, a component of the tectorial membrane. Knockout mice showed detachment of tectorial membrane from the cochlear epithelium. COL11A2 encode type XI collagen. The knockout mice have atypical and disorganized collagen fibrils of the tectorial membrane. COCH protein is ubiquitously present in the inner ear. It is expressed in cochlea as well as vestibular organs. COCH mutation exhibits symptoms of Meniere, including vertigo, tinnitus and a pressure feeling in the ear.
Mitochondrial Hearing Loss
In addition to nuclear genes some mitochondrial genes have been associated with genetic hearing loss. As mitochondria have a crucial function in nearly every cell, it is not unexpected that mitochondrial DNA mutations mainly cause multisystemic diseases, of which hearing impairment is often an additional symptom. Two different mitochondrial genes are involved. First, the 1555A->G mutation in the 12S rRNA gene leads to aminoglycoside-induced hearing loss and nonsyndromic hearing loss. The second gene is the tRNASer(UCN) gene, in which four different mutations are known to cause nonsyndromic hearing impairment.
A complete history should include prenatal, perinatal, postnatal, and family history. Especially in family history, it is important to inquire hearing loss in the first and second degree relatives, especially if the loss started before age 30. Consanguinity or common origin form ethnically isolated areas should increase suspicion of hereditary hearing loss. If there are a number of family members with hearing loss, constructing a pedigree is important.
Physical exam should focus on looking for features associated with congenital infection or syndromic hearing loss. Note hair color, the presence of a white forelock, facial shape, and skull shape. On eye examination, one should note the color of the iris, position of the medial and lateral canthum, intercanthal distance, cataracts, and any retinal findings. Examine the ear for preauricular pit, skin tags, shape and size of the pinna, or abnormality of EAC or TM. Examine the neck for branchial anomalies and thyroid enlargement and oral cavity for cleft. Thorough inspection of skin for areas of hyper or hypopigmentation and café-au-lait spots. Do a complete neurological exam including gait and balance to evaluate vestibular function.
In addition to audiogram, lab test should be ordered based on the history and physical exam. All children with hearing loss should have a urinalysis to assess for renal dysfunction. Other tests should be ordered as appropriate, for example, thyroid function test and EKG in suspected Pendred syndrome and Jervell and Lange-Neilsen syndrome. A CT scan of temporal bone is the radiological test of choice for evaluation of pediatric sensorineural hearing loss and should be considered in all patients with progressive hearing loss and craniofacial anomalies. The most common finding is dilated vestibular aqueduct, followed by Mondini malformation. Dilated vestibular aqueduct (DVA) suggests the diagnosis of Pendred syndrome, but may be found in branchio-oto-renal syndrome or in isolation. All patients with documented DVA should be investigated for Pendred syndrome.
Genetic counseling is an important part of evaluation. The counseling team usually consists of clinical/medical geneticist, genetic counselor, social worker, and psychologists. The risk and benefit of genetic testing should be explained to the family and consent should be obtained before genetic testing. Three genetic tests are available: GJB2, SLC26A4, and EYA1. Knowing the genetic cause of a person’s hearing loss can lead to improved decision about treatment and management. Genetic information can help predict whether the hearing loss will remain the same or whether it will worsen over time. Knowledge of the genetic cause is also helpful in determining what kind of damage has happened to the hearing system to cause the deafness. This is important because how the inner ear is damaged may affect whether a cochlear implant, or other hearing device, may help a patient. In addition, genetic testing can help determine if problems besides hearing loss may be present or may develop in the future. Genetic testing can also provide a deaf individual or the parents of deaf child information when making reproductive choices.
Since the identification of the first deafness gene in 1995 a respectable number of genes have been identified. As a consequence, the molecular knowledge of the processes responsible for hearing and balance and of the pathological mechanisms leading to hearing loss is expected to increase tremendously. But there is still a long way to go. Many genes remain to be identified. Moreover, the elucidation of the exact function of genes for which only a putative function was proposed, and of genes with an unknown function remains a great challenge.