Chapter 15: Otolaryngology: Head & Neck Surgery

From hearing loss or nasal hemorrhage (epistaxis) to endocrine surgery and expert management of acute airway emergencies, otolaryngology/head and neck surgery is a surgical subspecialty which focuses on the management of a wide range of disorders of the head and neck. As a colleague once put it, “Otolaryngology pretty much covers everything above the clavicles except the eyes, the brain, and the spinal cord.” The limits inherent to a single book chapter preclude covering such a broad field comprehensively. Therefore this chapter will present an overview of selected disease processes in otolaryngology that are of importance to the general surgeon in training. Other important parts of otolaryngology such as surgical endocrine disorders of the neck (eg, thyroid and parathyroid), facial plastic and reconstructive surgery, and facial skeletal trauma are covered in separate chapters.

Anatomy and Physiology

The external ear consists of two parts, the auricle (projecting from the lateral aspect of the head) and the external auditory canal (EAC) projecting medially to the tympanic membrane. Functioning as resonant amplifiers of sound energy, the concha of the auricle (Figure 15–1) has a resonance frequency of approximately 5 KHz, and the EAC has a resonance frequency of approximately 3.5 KHz. Combined, the external ear amplifies sound by approximately 10-15 dB in the 2-5 KHz range.

The tympanic membrane is positioned in an oblique plane, separating the EAC from the middle ear. It functions in transforming acoustic energy from sound waves to mechanical energy, which is transmitted via the ossicles—malleus, incus, and stapes to the oval window of the cochlea. The mechanics of the middle ear further amplify sound energy using two methods. First, the tympanic membrane is approximately 17 times larger than the footplate of the stapes; second, the ossicles act as a lever, providing a mechanical advantage of 1:1.3 from the tympanic membrane to the oval window. Combined, the result in a 25-30 dB gain in amplification.

The temporal bone houses the bony portion of the EAC, the middle, and the inner ear. The otic capsule of the inner ear is the hardest bone in the human body. Other important structures passing through or adjacent to the temporal bone include the carotid artery, the jugular vein, and the facial nerve (seventh cranial nerve). All of these structures are at risk for injury from temporal bone trauma.

The inner ear consists of the cochlea, which is both the auditory and vestibular sense organ. The vestibular system senses both linear acceleration (gravity) and angular acceleration (rotation). The hearing portion of the cochlea is a coiled tube that resembles a snail. Divided into three separate chambers, the scala vestibule and the scala tympani are filled with perilymph (similar in composition to extracellular fluid), while the scala media is filled with endolymph (similar in composition to intracellular fluid). The endolymph composition is maintained by Na+/K+ ATPase pumps within the stria vascularis found on the lateral walls of the scala media. These different chambers thus have different electrolyte composition, creating an electrical potential between the compartments. The sound energy, once transferred through the ossicles to the oval window of the cochlea, is coupled directly to the perilymph of the scala vestibule.

The resulting traveling wave is in the form of mechanical energy, which is then converted to electrical (neural) impulses within the scala media by the organ of Corti. The organ of Corti consists of inner hair cells (which are the sensory cells) and outer hair cells (which function as modulators of the inner hair cells), support cells and tectorial membrane. These nerve impulses produced by the inner hair cells are transmitted to the brainstem by the eighth cranial nerve, which travels from the cochlea through the internal auditory canal (IAC).

The vestibular system consists of the utricle, saccule, and three semicircular canals. Enveloped within the endolymphatic membrane, which is filled with endolymph, they are surrounded by perilymph, and then the very hard bone of the otic capsule. The utricle detects horizontal acceleration while the saccule detects vertical acceleration. Three semicircular canals, situated at right angles to each other, and paired with a semicircular canal on the opposite side of the head, detect angular acceleration.

Simplified, linear or angular motion causes their respective sensory cells to deflect, and the sensory cell body to depolarize. Depending on the direction, each vestibular apparatus will either increase or decrease the discharge rate relative to the basal rate, providing both direction, and speed of acceleration. Vestibular information is transmitted to the brainstem by the vestibular branch of the eighth cranial nerve.

Urgencies and Emergencies

A. Sudden Sensorineural Hearing Loss

The sudden onset of unilateral (or more uncommonly bilateral) hearing loss occurs at an annual incidence of 5-20 cases per 100,000 persons and can be an extremely unsettling experience for a patient. Most experts use a definition based on at least a 30 dB hearing loss occurring over 3 days or less. The causes of SSHL include viral infection (particularly herpes-family viruses), trauma, vascular compromise from thromboembolic phenomena or vasospasm, autoimmune disease, ototoxic (from chemotherapy, antibiotics, or salicylates), and congenital anatomic defects.

Prompt evaluation and initiation of treatment is critical, and may improve the prognosis for hearing improvement. Because of the multitude of potential mechanisms, a careful history should be performed. Often, the cause remains unknown. Patients may report an antecedent loud noise exposure such as an explosion (suggesting traumatic perilymphatic fistula), or upper respiratory tract infection symptoms (suggesting possible viral mechanism). Recent heart surgery or thromboembolic phenomenon could suggest a vascular etiology. Patients should be questioned about coincident imbalance or vertiginous symptoms, and tinnitus (ringing in the ears), indicative of vestibular as well as auditory pathology.

Formal pure-tone audiometry testing should be obtained. If the patient reports vestibular or balance symptoms, investigation of the vestibular system should be performed. Dix–Hallpike and Baranay maneuvers test for vertigo in response to specific position changes with respect to gravity. Simple tests of the auditory system (Weber and Rinne tests) can also be performed at the bedside using a 512 KHz tuning fork (Table 15–1). Pneumatic otoscopy should be performed, to evaluate for fistula sign (vertigo on pneumotoscopy insufflation).

Workup should include evaluation for a vestibular schwannoma, as this represents approximately 1%-3% of the sudden sensorineural hearing loss (SSHL). This can be done by the auditory evoked brainstem response (ABR), or a magnetic resonance imaging (MRI) with contrast using IAC protocols. Some experts advocate a less expensive screening MRI without contrast as an acceptable first step, followed by more definitive evaluations if any abnormalities are present. Laboratory studies should include a CBC with differential, ESR, PT/PTT, and cochlear antibodies. Additional studies that may be helpful include syphilis testing (either MHA-TP or FTA-Abs), and thyroid function tests. If there is a family history of sudden hearing loss, a computed tomography (CT) scan looking for enlarged vestibular aqueduct can be obtained.

If the workup reveals a cause for the SSHL, this should be addressed. Unfortunately, the majority remain idiopathic. In this case, unless contraindicated due to comorbidities, initial treatment should be started empirically. Oral corticosteroids, such as a prednisone taper (60 mg daily for 9 days then tapering over 5 days), and antiviral medication (such as acyclovir) should be administered for a minimum of 2 weeks. Other treatments such as hyperbaric oxygen, carbogen inhalation, anticoagulation, and diuretics have been proposed for the treatment of SSHL but results of these therapies have been inconclusive.

The prognosis for spontaneous recovery of function is hopeful, and approximately 60% of patients will recover full or partial hearing. With corticosteroid treatment, this may increase to 80% according to some studies. The addition of acyclovir has not been shown to be effective, but some experts recommend it due to low side effects and a plausible mechanism of action.

B. Acute Facial Nerve Paralysis

The seventh cranial nerve (facial nerve) innervates the muscles of facial expression (as well as special motor afferents to the parotid and lacrimal glands). A careful history will help define the onset of paralysis (eg, acute deterioration over less than 2-3 days, or gradual decline). Antecedent events (temporal bone trauma, acute otitis media [AOM], hearing loss or imbalance, and recent viral illness) should be elicited and can help guide further workup and management.

Potential causes of acute facial nerve paralysis are numerous; however over 50% are idiopathic and termed “Bell Palsy.” Note that Bell Palsy is a diagnosis of exclusion, and therefore a focused but complete workup must be performed. Fully 20% of facial nerve paralyses are caused by trauma. Other common causes (but by no means an inclusive list) include herpes zoster oticus (Ramsey Hunt syndrome), complications from otitis media and mastoiditis, Lyme disease, cholesteatoma, and neoplasm.

A complete physical examination with emphasis on the neurologic system and cranial nerves is critical. An important consideration in evaluating the patient with acute facial nerve paralysis is distinguishing between central and peripheral lesions. In central lesions, there is sparing of forehead elevation on the affected side due to decussating fibers from the contralateral side. In peripheral lesions the fibers have already crossed and there is no forehead sparing. The degree of facial nerve functional loss should be documented and has important bearing on prognosis. Several grading systems have been proposed; the House–Brackmann scale is the most widely used and is presented in Table 15–2. Facial nerve function is graded from I (normal) to VI (total paralysis) for each side.

Evaluation should include pure-tone audiometry and electrophysiologic testing. A high-resolution CT of the temporal bone is the imaging study of choice for evaluating bone changes (from mastoiditis, temporal bone trauma, cholesteatoma, or neoplasm) whereas MRI with contrast is helpful when inflammation (eg, herpes zoster oticus) or neoplasm affecting the nerve is suspected. Laboratory studies should include a CBC with differential and ESR or CRP. If these are clinically suspected, autoimmune serologies and Lyme titers can be obtained.

If the facial nerve paralysis is caused by traumatic injury, management depends on the onset of paralysis; immediate and complete paralysis will often benefit from surgical decompression. In delayed or partial dysfunction, spontaneous recovery is likely and surgical intervention may not be beneficial. For idiopathic paralysis (Bell Palsy), initial medical therapy is targeted to reducing inflammation and targeting a possible viral etiology. A corticosteroid course should be initiated, either prednisone or prednisolone. Previous recommendations for the use of acyclovir (or similar antiherpetic analog) have recently been questioned. A large-scale randomized double blind study by Sullivan et al demonstrated no benefit for acyclovir, but significant benefit for prednisolone in the treatment of Bell Palsy. If the paralysis progresses, surgical decompression may be beneficial. Most experts also argue a significant benefit to performing surgical decompression of the facial nerve if a patient progresses to severe paralysis, provided the decompression is performed within 14 days of the onset of paralysis. Severe paralysis is usually defined as developing greater than 90% degeneration on electroneurography (ENoG) and lacking voluntary motor potentials on EMG.

C. Foreign Body in the External Auditory Canal–Button Batteries

The vast majority of cases of foreign bodies in the EAC occur in children, or mentally impaired adults. Many otologic foreign bodies, if carefully selected, are actually amenable to removal under direct visualization in an emergency department or primary care office setting. In a large series of over 600 cases of ear foreign bodies, Shulze et al found an overall 77% success rate for the removal under direct visualization by emergency physicians. It is important to note that the majority of those foreign bodies successfully removed in this manner fit the category of “soft, irregular” material such as paper or cotton, and “pliable or rubber-like” such as silly putty or erasers. Success with hard objects, and especially spherical objects such as plastic beads was markedly lower. Thus, an argument can be made for a single attempt by the pediatrician or emergency physician under direct visualization, if the object meets the former criteria. This should be tempered, however, with the understanding that complication rates are much higher with removal under direct vision. Complications of foreign body removal most commonly include canal wall lacerations (47%), with tympanic membrane perforations less common (4%). More serious complications such as ossicular chain injury and oval window perforation are possible but rare. When initial foreign body removal is performed instead by the otolaryngologist using binocular otomicroscopy either in the office or operative setting, the complication rate is quite low (6.3%).

Foreign body removal using binocular otomicroscopy is the primary method used by most otolaryngologists. Specific techniques depend on the characteristics of the foreign body. Objects with sharp edges can often be grasped with alligator or duck-bill forceps. Soft objects are often amenable to removal with otologic suction. Removal of spherical and hard, irregular objects requires more finesse. In these cases a 90-degree probe is invaluable. The probe is carefully guided behind the object under binocular otomicroscopic visualization. Then the probe is rotated along its axis to bring the end behind the object. The object is then guided out of the EAC. Another special case is an insect within the EAC, most commonly cockroaches. This can be quite alarming to the patient if the insect is still alive. The proximity of the insect to the tympanic membrane translates movement of the insect to distressingly loud perceived sound levels. In such cases the external ear canal can be gently irrigated with either mineral oil or lidocaine to suffocate the insect followed by prompt removal.

Some non-otolaryngologists advocate the use of gentle irrigation to attempt dislodgement of EAC foreign bodies. This technique can be used successfully, but requires caution. First, if there is any suspicion of tympanic membrane perforation, irrigation is contraindicated as it could flush debris into the middle ear space. Second, if the foreign body is composed of vegetable material (such as a popcorn kernel) irrigation should be avoided. If the object is not successfully flushed out, subsequent swelling of the vegetable foreign body can result in extreme pain as the EAC skin is compressed against the bony canal. Removal, once this occurs can be problematic, and may require general anesthesia and the use of an operating microscope. For this reason, otologic medications are also contraindicated in vegetable material otologic foreign bodies. The third condition where irrigation (and otologic medications) is specifically contraindicated is in the case of EAC button-batteries.

While the previously described foreign bodies can be managed on an outpatient basis, the finding of a button battery in the EAC is considered an emergent situation, requiring urgent removal by an otolaryngologist. If left in place for any length of time, batteries in the EAC can result in severe complications. In an early description of this problem by Kavanagh et al, 100% of patients experienced multiple, serious sequelae. These included tympanic membrane perforation or total destruction (75%), marked dermal destruction with bone exposure (88%), impairment of hearing (38%), ossicular chain erosion (25%), and even facial nerve paralysis (13%). Both leakage of corrosive battery acid and electrical current discharge resulting in chlorine gas and sodium hydroxide by electrolysis have been hypothesized to contribute to the destructive effects of button batteries. Removal is accomplished under binocular otomicroscopy in the operating room, and sometimes requires piecemeal removal of the battery. Following removal of the battery, the EAC should be flushed with copious amounts of saline, and careful inspection of the external canal and tympanic membrane should be performed.

Disorders and Diseases

A. Otitis Externa

Otitis externa is an infection of the EAC, usually from bacterial species such as Pseudomonas, Proteus, Klebsiella, Streptococcus, and Enterobacter. Evaluation and management of a patient with suspected otitis externa includes a complete history and physical examination with emphasis on the otologic examination. Patients will often give a history of recent water exposure to the external ear, such as from swimming, or other predisposing factors such as chronic hearing aid use. The external pinna should be manipulated gently. In otitis media this should not elicit pain, whereas for patients with otitis externa movement of the pinna is extremely painful. On handheld otoscopy the canal wall will appear edematous and erythematous. Sometimes the edema is severe enough the tympanic membrane cannot be visualized. In this case, insertion of a Pope otowick is indicated to carry ototopical medications the length of the external ear canal past the obstructed site. Severe cases may require frequent suction debridement under microscopic visualization. Patients should be placed on ototopical antibiotic drops containing a topical corticosteroid, such as ciprofloxacin 0.3%/dexamethasone 0.1% suspension (Ciprodex). Some (largely nonotolaryngologist) physicians allow the use of hydrocortisone 1%/polymyxin/neomycin (Cortisporin) as an alternative. This is not recommended for several reasons. First, several authors have demonstrated up to 10% risk of contact dermatitis with polymyxin/neomycin ototopical drops. Second, if a possibility of tympanic membrane perforation exists, these drops carry the possibility of ototoxicity, according to laboratory studies in animals; they are not approved for the middle ear use (unlike fluoroquinolones). Additionally, some studies have demonstrated faster pain relief with Ciprodex compared to Cortisporin. It is essential to note that failure to respond to appropriate therapy within 48-72 hours should prompt the clinician to reassess the patient to confirm a diagnosis of otitis externa.

Of note, patients with a history of diabetes (or any immunocompromising condition) should demand more aggressive therapy targeted toward Pseudomonas. This will usually involve an ototopical fluoroquinolone, ototopical corticosteroid, and oral or IV fluoroquinolone therapy. In the past, this subset of otitis externa carried the misnomer “malignant otitis externa” due to the high mortality rate even with surgical debridement and antibiotics. Modern antibiotic therapy and earlier diagnoses and intervention have dramatically improved outcomes, and mortality from this disease is now relatively rare.

Otomycosis is otitis externa due to a fungal infection, usually due to Aspergillus or Candida species. On otoscopy, there is usually far less edema and erythema than with bacterial infection. The use of ototopical antibiotic solutions will not improve these patients, and usually worsens their condition. Otomycosis can be notoriously difficult to treat, but many cases do respond to suction debridement followed by acidic eardrops and a topical corticosteroid. A commonly used topical preparation is acetic acid 2% with hydrocortisone 1% (Vosol HC). Topical antifungals such as nystatin or amphotericin B are also available but use of these should be reserved for difficult cases under the care of an otolaryngologist.

B. Otitis Media

Classification of common middle ear pathology is often poorly understood by nonotolaryngologists. There are multiple disease processes of the middle ear which include the root term “otitis media.” Additionally, the acronyms used to represent diseases of abnormal middle ear fluid are quite similar.

The first condition, AOM, represents what is commonly referred to as “a middle ear infection.” The typical patient is a young child, with history of upper respiratory tract symptoms, fever, and pulling at one ear. Handheld otoscopy will reveal a bulging, erythematous tympanic membrane. Unlike otitis externa (above) there is no pain with manipulation of the auricle. Several studies from Europe have demonstrated that the majority of AOM cases resolve spontaneously without intervention. Nevertheless, in the United States most parents would be unhappy with a decision not to treat, and antibiotic therapy for AOM is routine. Usual pathogens include bacteria such as Streptococcus, Haemophilus, and Moraxella. The latter two are often resistant to penicillins and thus treatment with amoxicillin may not clear the infection; thus many practitioners recommend a second-generation cephalosporin. Failure to respond to these agents often necessitates a second-line antibiotic such as amoxicillin-clavulanic acid (Augmentin).

Otitis media with effusion (OME) is defined as fluid in the middle ear but no active signs of an infection. OME often results from eustachian tube dysfunction, which predisposes to accumulation of a sterile serous fluid in the middle ear cleft that does not clear. OME is common in young children, with a prevalence approaching 30% according to some reports. Patients with OME present with complaints of muffled or decreased hearing. Examination reveals fluid in the middle ear cleft. Chronic serous otitis media (CSOM) results when middle ear fluid after an episode of AOM fails to clear after a reasonable time span (4-6 weeks).

The most common surgical intervention to treat these problems is myringotomy and tympanostomy (M&T), so-called “ventilation tube” placement. Basic indications for M&T include multiple episodes of AOM (four episodes in 6 months, or six episodes in 12 months), CSOM with hearing impairment present for 3 months or longer, or presence of complications of AOM. Some authors have advocated for a more stratified approach to indications, where patients presenting with problems at an early age warrant surgical intervention with less stringent criteria.

C. Vestibular Schwannoma

Vestibular schwannomas (also sometimes referred to by the misnomer “acoustic neuromas”) represent a nonmalignant but neoplastic proliferation of Schwann cells ensheathing the eighth cranial nerve. These tumors represent nearly 10% of all intracranial tumors, and usually present with a unilateral high-frequency sensorineural hearing loss, followed later by development of imbalance symptoms. Even a mild degree of hearing loss may be misleading, as sensory processing (as evidenced by speech discrimination scores) are often more impaired than pure-tone averages would predict. Tinnitus and true vertigo are less common symptoms. Interestingly, these tumors arise more often from the vestibular division than the auditory division of cranial nerve eight. The symptoms associated with vestibular schwannomas are associated with compressive effects from the neoplastic growth. In the case of larger tumors, patients can sometimes present with facial nerve weakness. One important syndrome associated with vestibular schwannomas is neurofibromatosis-2 (NF2). Patients with NF2 can present with bilateral vestibular schwannomas, and consideration of hearing preservation strategies thus is extremely important for these patients.

Presumptive diagnosis of vestibular schwannoma is usually made by MRI with gadolinium. Schwannomas enhance brightly on T1 or T2 weighted images with gadolinium. Other diagnostic tests can include auditory brainstem response (ABR), electronystagmography (ENG). All patients with suspected vestibular schwannomas should have audiometric testing (pure-tone averages and speech discrimination scores) performed.

Management of vestibular schwannomas remains controversial. Many authors advocate that small schwannomas confined to the IAC can be successfully watched via serial imaging. Tumors demonstrating no growth (<2 mm) can continue to be watched, while most authors would argue for intervention if more than 2 mm growth occurs. If intervention is elected, treatment can be microsurgical (often involving combined neurosurgical and otologist collaboration) or by stereotactic radiosurgery (gamma knife) for small tumors. The goal of treatment is eradication of the tumor while preserving hearing and facial nerve function when possible. Preservation of facial nerve function (House-Brackman grade 2 or better) is generally successful in up to 70% of patients, regardless of surgical approach used.

D. Benign Paroxysmal Positional Vertigo

Dizziness is an extremely common phenomenon, and has been estimated to affect as many as 30% of patients. An important distinction should be made at this point between various sensations commonly described by patients as “dizziness.” For patients describing dizziness symptoms, an attempt to elicit a more accurate description should always be sought. Vertigo is defined as the illusion of rotation. This should be distinguished from sensations of imbalance or unsteadiness, or of almost losing consciousness (presyncope).

Benign paroxysmal positional vertigo (BPPV) is the most common cause of acute-onset vertigo. Patients will describe sudden onset of intense vertigo lasting seconds rather than minutes, usually brought on by changes in head or body position relative to gravity. There is sometimes an associated history of head trauma. The etiology is thought to be due to dislocation of micro-crystals of calcium hydroxyapatite (otoconia) from the vestibule into the posterior semicircular canal. Certain head movements cause the otoconia to abnormally trigger copular deflection and thus elicit imbalanced vestibular input to the brainstem triggering intense vertigo. Diagnosis of BPPV can be made by positional testing, such as the Dix–Hallpike maneuver. In this test, the patient’s head is turned to one side and the patient is laid into a recumbent position with the head maintained in the rotated position. Elicitation of vertigo, often accompanied by the expected rotatory nystagmus, essentially confirms the diagnosis. Treatment consists of directed repositioning techniques such as the Epley maneuver. These maneuvers are designed to rotate the otoconia through the semicircular canal, depositing them back in a more physiologic location within the vestibule. Often several treatments are required, and patients can be instructed in the self-application of these maneuvers.

E. Méniére Disease

Méniére disease is characterized by waxing and waning sensorineural hearing loss (typically low frequency more than high), episodes of vertigo, sensation of aural fullness, and tinnitus. Hearing loss typically follows an episodic but slowly progressive course, with times of worse hearing followed by partial recovery. The hearing loss (and vestibular dysfunction) is most commonly unilateral, although bilateral disease can develop. The vertigo attacks associated with Méniére disease can be debilitating, and are often accompanied by nausea, vomiting, and inability to perform normal activities.

Diagnosis of Méniére disease is made based on clinical presence of the tetrad of symptoms, combined with evidence of hearing loss and vestibular dysfunction. The episodic nature of the disease is an important characteristic; a single episode of hearing loss and vertigo should not prompt a diagnosis of Méniére disease. In this case, viral labyrinthitis is a more likely culprit.

First described in 1861 by Prosper Méniére, the pathogenesis of Méniére disease remains essentially unknown. It is thought to relate to dilation of the membranous labyrinth, possibly from dysfunction within the endolymphatic sac. Anatomic cadaver studies have demonstrated endolymphatic hydrops (swelling of the scala media and endolymphatic sac) in patients with Méniére disease. Unfortunately, these anatomic changes have also been demonstrated in presumably normal (or at least asymptomatic) patients.

The mainstay of treatment for Méniére disease remains medical therapy. Patients are typically begun on a low-salt diet initially. Patients are also instructed to avoid caffeine, nicotine, and alcohol. Diuretics can be added, along with vestibular suppressants such as diazepam. Antihistamines (meclizine, dimenhydrinate, etc) have also demonstrated benefit in ameliorating vertigo symptoms associated with Méniére disease.

Patients suffering from severe, incapacitating vertigo and failing medical therapy can be offered several surgical interventions. Many authors advocate unilateral chemical ablation of the vestibular system. This is most often accomplished by transtympanic injection of gentamycin, and while efficacious for vertigo resolution is associated with sensorineural hearing loss in up to 25% of patients. For patients with intact hearing, many experts feel endolymphatic sac decompression and shunting can offer significant relief with preservation of hearing, although this remains unproven in randomized controlled trials. Similarly unproven, vestibular neurectomy (selective sectioning of the vestibular branch of the eighth cranial nerve) may conserve hearing and provide vertigo relief. For patients with severe hearing loss and vertigo, a total transmastoid labyrinthectomy relieves vertigo in over 90% of patients, at the cost of complete hearing loss on the affected side.

F. Cholesteatoma

A cholesteatoma is a cyst-like, expansile lesion of the temporal bone consisting of stratified squamous epithelium and trapped desquamated keratin. It occurs in the pneumatized temporal bone, most commonly the middle ear and mastoid (Figure 15–2). There are two types of cholesteatoma (acquired and congenital) with the former being the most common. Acquired cholesteatomas arise from either retraction pockets within the tympanic membrane, or secondarily from a tympanic membrane perforation. Congenital cholesteatomas are thought to arise from epithelial cell rests that fail to undergo apoptosis during development. Whatever the origin, once formed cholesteatomas behave in a locally destructive manner. Bone erosion is common, especially of the ossicular chain but also potentially the bone surrounding the inner ear (the otic capsule). If left untreated cholesteatomas can even invade intracranially.

Early cholesteatomas have few if any symptoms, and usually start with a slowly progressive hearing loss. If an infection develops in a cholesteatoma, a foul otorrhea will develop, and this sometimes is the presenting symptom. If a cholesteatoma is suspected, careful inspection under binocular otomicroscopy is essential. All debris must be removed to allow a full visualization of the entire visible portion of the tympanic membrane. Cholesteatomas will appear as a whitish keratin mass. A pneumatic otoscopy test is essential; if vertigo is elicited the surgeon must suspect erosion into the inner ear structures.

The treatment for cholesteatoma is surgery, usually involving removal of the mastoid air cell septations by otologic drill, exposing the middle ear space. This accomplishes two goals—providing safe visualization and access and removal of all cholesteatoma tissue. This procedure is performed under careful microscopic visualization as many important structures (such as the facial nerve and the inner ear) should be preserved. The primary goal of surgery is creation of a safe, dry ear. All other considerations, including preservation of hearing, take second place.

Anatomy and Physiology

The nose and paranasal sinuses serve to warm, filter, and humidify inspired air, to modulate vocalizations and speech, and provide for the sense of smell. The external nose consists of soft tissue and skin resting on a largely cartilaginous framework. The internal nose (nasal cavity) begins at the nasal vestibule anteriorly and extends posteriorly to the choana (which forms the boundary between the nasal cavity and the nasopharynx).

The nasal cavity (Figure 15–3) is divided in the sagittal plane in largely symmetric halves by the nasal septum. These cavities are partially filled by the three turbinates (superior, middle, and inferior, and occasionally a fourth supreme turbinate), which arise from the lateral nasal wall. The spaces below each turbinate are called meati (eg, superior meatus, middle meatus, and inferior meatus). These meati are important for localizing the outflow tracts of the various paranasal sinuses, which drain in a characteristic pattern. The nasolacrimal duct drains into the inferior meatus. The maxillary, frontal, and anterior ethmoid sinuses all drain into the middle meatus. The sphenoid sinus and posterior ethmoids drain into the superior meatus. Additionally, the olfactory nerve endings (the end organ for the sense of smell) are located on the superior nasal septum and superior turbinate mucosa.

The paranasal sinuses consist of hollow cavities that derive from pneumatization into the frontal, ethmoid, sphenoid, and maxillary bones of the craniofacial skeleton. They are lined with respiratory epithelium (pseudostratified ciliated columnar epithelium), which serves to circulate and drain mucous along with entrapped particulate matter. They are normally air-filled but can become fluid-filled if the ostia become obstructed by inflammation, anatomic problems, or disease process.

Urgencies and Emergencies

A. Nasal Foreign Body

As is the case for foreign bodies of the external ear, nasal foreign bodies are typically encountered in children. Typical presentation is a young child with several day history of unilateral foul rhinorrhea. Oftentimes, the offending foreign body was inserted days to weeks before symptom onset, and the patient or parents may not recall a specific event precipitating the symptoms.

Offending foreign bodies can include vegetative, inert (plastic/metal) material and button batteries. Unlike external ear foreign bodies, nasal foreign bodies should always be treated as relative urgencies regardless of the type of object present. This is because of the anatomic relationship of the nose to the upper airway—if the object becomes dislodged, it could easily become an airway foreign body (a true emergency).

For this reason, many otolaryngologists advocate recommend removal of all but the most anteriorly placed nasal foreign bodies under general anesthesia using endoscopic visualization. The threshold for performing retrieval under sedation should likewise be low. In our practice, nasal foreign bodies confined to the nasal vestibule, or easily visible by anterior rhinoscopy can be safely removed without sedation in a cooperative patient. Young children, superior or posterior location or difficult visualization all result in removal in the operating room with endoscopic visualization. Another advantage to removing nasal foreign bodies in this manner is that postremoval inspection of the nasal cavity is easily performed. Oftentimes the nasal mucosa can be significantly inflamed and placement of absorbable material can prevent unwanted adhesions between opposing mucosal surfaces.

Nasal button batteries present a similar situation to external ear button batteries (see External Ear Foreign Body, above). The potential for extensive tissue damage from leakage of acid and electrical current discharge usually necessitates removal under general anesthesia. Following removal, extensive flushing with normal saline and careful inspection of the nasal cavity should be performed.

B. Invasive Fungal Sinusitis

Invasive fungal sinusitis is almost always encountered in immunocompromised patients, most often patients undergoing chemotherapy for malignancy, or poorly controlled diabetics. It is caused by uncontrolled infiltrative growth of usually non-pathogenic fungal organisms. Offending agents are ubiquitous in the environment, and are commonly found in the nasal secretions of healthy normal patients. The two most common fungi are Aspergillus and Rhizopus species. The latter is more aggressive and is termed Mucormycosis. Rhizopus tends to preferentially grow in acidic environments, and is thus found more often in the setting of diabetic ketoacidosis. Even with early diagnosis and maximal surgical therapy, and modern antifungal agents, the disease still carries a significant mortality rate. This varies depending on causative agent, from approximately 10% (Aspergillus) to 30% (Rhizopus).

The index of suspicion for invasive fungal sinusitis must be high for any immunocompromised patient, as the symptoms can be subtle and the disease rapidly progressive. Patients are usually ill appearing and complain of facial pain, headache, nasal discharge, and may have mental status changes. Careful inspection of the face, oral cavity, and nasal cavity is mandatory. Dark ulcers (Figure 15–4) may be noted on the anterior face of the middle turbinate, inferior turbinate, lateral nasal wall, septum, or palate. Cranial nerve deficits may be noted in later stages of the disease.

If invasive fungal sinusitis is suspected, biopsies should be taken of the suspicious areas and sent for immediate pathologic examination. The pathologist should be alerted that invasive fungal sinusitis is suspected, so that special fungal stains can be performed. A worrisome finding is lack of bleeding at the biopsy site, as this may signify retrograde tissue infarction secondary to angioinvasive fungus.

Once invasive fungal sinusitis is confirmed, therapy consists of rapid reversal of immunosupression followed by aggressive surgical debridement of all necrotic tissue (down to healthy, bleeding tissue) and systemic antifungal therapy. Two commonly used agents are voriconazole and amphotericin B. Usually multiple surgical debridements are required. Typically, patients do not recover unless the underlying immunosuppression is resolved; for example absolute neutrophil counts greater than 500 or diabetic ketoacidosis rapidly reversed. Adjuvant therapy with donor neutrophil transfusion is an emerging treatment addition especially in cases failing to respond to conventional treatment. A multidisciplinary team including otolaryngology, hematology/oncology, and infectious disease specialists should closely follow these patients. Many patients develop metastatic fungal infections and can develop necrotic cavitation at distant sites. This often occurs as the patient’s immune system recovers, and can result in lethal pulmonary hemorrhage, strokes, and other serious systemic sequelae.

Invasive fungal sinusitis should not be confused with other forms of fungal disease of the paranasal sinuses. In allergic fungal sinusitis, chronic host immunologic/inflammatory response to noninvasive fungal elements results in tissue eosinophilia, nasal polyps (hyperplastic growth that obstructs normal drainage and airflow), and bony remodeling.

C. Epistaxis

Epistaxis most commonly arises from the anterior portion of the nasal septum, referred to as Little’s area. In this region, blood vessels derived from both the internal and external carotid anastomose to form Kiesselbach plexus. This rich blood supply can result in quite dramatic amounts of bleeding which can be distressing for the patient. Fortunately, many of these episodes respond to simple external pressure (pinching the anterior external nose) for 10 minutes.

The etiology of epistaxis is primarily related to disruption of nasal mucosa, thus exposing small blood vessels that can rupture. In children this often relates to nose picking. In adults, the etiology often relates to turbulent nasal airflow, such as from a deviated nasal septum. Other predisposing factors include drying of the nasal mucosa, and hypertension. The latter is particularly important in the acute management of epistaxis; often the bleeding will not be controllable until the accompanying hypertension is dealt with.

Epistaxis not responding to conservative management may require intranasal tamponade. Several different packing devices, including petrolatum gauze, thrombin containing collagen products, sponges, and inflatable balloons can be used to perform anterior packing. Anterior bleeding from Little’s area can also sometimes be controlled with topical silver nitrate chemical cauterization, usually under endoscopic visualization.

Another common source of epistaxis is bleeding from branches of the anterior ethmoid or sphenopalatine artery. Epistaxis from these often requires posterior packing (such as with Foley catheter occlusion of the choanae and complete obliteration of the nasal airspace with gauze packing). Patients requiring posterior packing should be hospitalized and placed on pulse oximetry. Of note, nasal packing materials should be covered with topical antibiotics before placement, and all patients with intranasal packing should be placed on antistaphylococcal antibiotics to prevent possible toxic shock syndrome. Epistaxis recalcitrant to anterior and posterior packing can be managed by surgical ligation of the offending arterial supply (internal maxillary, sphenopalatine, and anterior ethmoid) or by arterial embolization by an interventional radiologist.

Chronic medical management of patients predisposed to epistaxis includes management of hypertension, and promoting moist nasal mucosa. Patients are commonly placed on two to three times daily application of nasal saline spray, and topical petroleum jelly or antibiotic ointment to the anterior septum. Patients with severe, recurrent epistaxis should be evaluated for possible systemic disease (such as hereditary hemorrhagic telangiectasia, Wegener granulomatosis, etc).

Disorders and Diseases

A. Acute Rhinosinusitis

Rhinosinusitis refers to inflammation of the mucosal lining of the nose and paranasal sinuses. Acute rhinosinusitis is present for less than 3 weeks, and is usually precipitated by a viral upper respiratory tract infection. It is important to emphasize that only a minority of cases of acute rhinosinusitis become complicated by bacterial superinfection (0.5%-2%). Similarly, change in color of nasal discharge is not a specific sign of bacterial rhinosinusitis.

Symptoms initially reflect the precipitating upper respiratory viral infection (cough, sneezing, fever, nasal congestion, facial pain/pressure, rhinorrhea, and sore throat) followed by development of rhinosinusitis symptoms. A set of diagnostic symptoms for rhinosinusitis (both acute and chronic) have been established. Patients must have two major, or one major and two minor criteria. These criteria are outlined in Table 15–3.

Acute bacterial rhinosinusitis should be suspected if symptoms persist after 10 days, or worsen within 10 days after an initial improvement. Physical examination may reveal purulent nasal secretions, nasal mucosal erythema, and tenderness overlying the sinuses. Nasal endoscopy (with middle meatal cultures) can be very useful, and visualization of purulent secretions emanating from the osteomeatal complex should increase suspicion for bacterial rhinosinusitis.

Treatment of acute rhinosinusitis is largely conservative in nature. Nasal saline lavage helps to eliminate excess mucous and inflammatory mediators, and restore mucocilliary clearance. Topical decongestants (such as oxymetazoline) can help reduce mucosal edema and restore sinus ostia drainage. Use should be restricted to 3 days, as tachyphylaxis and dependence can result from overuse of topical decongestants. Mucolytics such as guaifenesin can help thin mucous secretions, allowing easier mucocilliary transport. The only therapy tested and confirmed by placebo-controlled trials is intranasal topical steroids (such as mometasone or flunisolide). Topical steroids have been demonstrated to reduce time to symptom resolution, both for bacterial and nonbacterial acute rhinosinusitis. It is important to note that antihistamines have no proven benefit in acute rhinosinusitis, and may actually cause symptom exacerbation by drying mucous secretions. Antibiotics should be reserved for patients suspected of having acute bacterial rhinosinusitis. Current recommendations specify first-line antibiotic therapy should consist of amoxicillin. After 7 days, if a patient fails to improve clinically a broader-spectrum antibiotic such as a fluoroquinolone, trimethoprim/sulfamethoxazole, azithromycin, or amoxicillin/clavulanic acid can be tried. All of these have greater than 80% efficacy in clearing acute bacterial rhinosinusitis.

Patients experiencing multiple episodes of acute rhinosinusitis should be evaluated carefully for predisposing conditions. These may include anatomic obstruction (which may be relieved by septoplasty and/or functional endoscopic sinus surgery), underlying impairment of mucociliary clearance (such as from immotile cilia/Kartagener syndrome, or cystic fibrosis), or immune system dysfunction. Complications of acute bacterial rhinosinusitis can include orbital cellulitis and subperiosteal abscess formation, meningitis, and cavernous sinus thrombosis.

B. Chronic Rhinosinusitis

Chronic rhinosinusitis is extremely common, and affects between 2% and 15% of people in the United States. It is defined as the presence of rhinosinusitis symptoms for greater than 12 weeks (major criteria from Table 15–3) combined with evidence of inflammation. The latter can include findings of purulent mucous in the middle meatus or ethmoid region, nasal polyps or polypoid degeneration of the nasal mucosa. Radiographic findings (Figure 15–5) can also document inflammation, most commonly by CT. Findings can include diffuse mucosal thickening, chronic bony remodeling, and sinus opacification.

Management of chronic rhinosinusitis is primarily by topical and systemic medication. Most otolaryngologists place chronic rhinosinusitis patients on topical nasal steroids for at least 1 month. At this time if symptoms persist, a CT scan may be useful in demonstrating any anatomic abnormalities that may be amenable to surgical correction. Surgery is aimed at removing the obstruction to natural mucous flow from the paranasal sinuses, and most patients will continue to require medication after surgery to prevent return of symptoms.

The etiology of chronic rhinosinusitis is not fully elucidated. Most otolaryngologists feel that there are several disease processes currently described together under the generic heading chronic rhinosinusitis. Tissue eosinophilia may play an important role in differentiating these groups; current molecular evidence supports this distinction. Future research will undoubtedly change our understanding of these disease processes and how they are managed.

C. Rhinitis Medicamentosa

As mentioned above, extended use of topical nasal decongestants (such as oxymetazoline) or other vasoconstrictors (such as intranasal cocaine) can lead to tachyphylaxis and mucosal dependence. The resulting severe mucosal edema, hyperemia, and nasal obstruction are termed rhinitis medicamentosa. Patients will report daily topical vasoconstrictor/decongestant use, and absolute dependence on these medications for any appreciable nasal airflow. It is not unusual for patients suffering from rhinitis medicamentosa to carry their topical vasoconstrictor medication with them due to their frequent use. This is often a telltale sign of dependence. On physical examination the nasal mucosa will be thickened, erythematous and edematous, and lack appreciable decongestion on topical decongestant application.

Treatment requires complete cessation of the offending agent. Patients should be started on nasal saline lavage and nasal topical steroids. Oral decongestants and a course of oral corticosteroids may help hasten symptom resolution and increase patient compliance. Resolution usually takes 3-4 weeks, and may require much more time in the case of long-term vasoconstrictor use, or intranasal cocaine abuse.

Complications of untreated rhinitis medicamentosa include poor healing after nasal surgery, septal perforation, and formation of synechiae. For this reason, it is important to recognize and treat rhinitis medicamentosa preoperatively, before embarking on any nasal surgical treatment.

Anatomy and Physiology

The oral cavity is bounded anteriorly by the vermilion border of the lips, and posteriorly by the anterior pillars of the palatine tonsils. The superior aspect of the oral cavity includes the hard and soft palate, and inferiorly includes the lingual mucosa and the anterior two-thirds of the tongue. This region of the tongue is bounded by the circumvallate papilla, which lie along the sulcus terminalis and separate the oral tongue from the base of tongue (part of the oropharynx).

The pharynx connects the nasal and oral cavities to the esophagus and larynx. It consists of three segments—the nasopharynx, the oropharynx, and the hypopharynx (Figure 15–6). The nasopharynx begins as an extension from the posterior aspect of the nasal cavity and extends from the nasal choana to the soft palate. The oropharynx extends from the soft palate to the level of the hyoid bone, and is bounded laterally by the tonsillar pillars (the palatoglossal and palatopharyngeal arches). It includes the base of tongue, lateral/posterior pharyngeal wall, and tonsillar fossae. The hypopharynx extends from the level of the hyoid bone to the inferior aspect of the cricoid cartilage, and includes the pyriform sinuses, postcricoid region, and posterior hypopharyngeal wall.

The primary functions of the oral cavity are related to chewing and swallowing (mastication and deglutition) and shaping of phonatory vibrations to produce intelligible speech. Taste buds on the dorsum of the tongue are responsible for basic taste perception—sweet, salty, bitter, and sour. This sensory information is transmitted to the facial nerve via the chorda tympani nerve for the anterior two-thirds of the tongue. General sensation of the tongue is carried by the lingual nerve. All sensory information from the posterior one-third of the tongue is carried by the glossopharyngeal nerve. Complex nuances of taste are mediated by olfactory receptors in the superior-most aspect of the nasal cavity and are not directly related to the tongue or oral cavity.

The tongue has four pairs of intrinsic muscles, which interdigitate throughout the tongue. These muscles act to lengthen or shorten the tongue, curl the apex and edges, and flatten or round the dorsal surface. The intrinsic tongue muscles originate and insert within the tongue itself. Extrinsic tongue muscles (genioglossus, hyoglossus, styloglossus, and palatoglossus) also act to protrude, depress, elevate, and retract the tongue. All motor function of the tongue is mediated by cranial nerve XII (the hypoglossal nerve).

The act of swallowing, or deglutition, is complex and consists of three main phases—oral, pharyngeal, and esophageal. The oral phase is under voluntary control, while the remaining phases proceed under reflex control. The oral phase of swallowing consists of preparation of the food bolus by mastication to soften and shape the bolus. Following this, oral transport ensues and the food bolus is transported to the posterior tongue. The anterior tongue then elevates against the hard palate, contracts and propels the bolus to the oropharynx. Simultaneously, the nasopharnyx is sealed off preventing nasal regurgitation. In the pharyngeal phase, several complex actions occur which elevate the larynx, temporarily halt respirations and protect the airway from aspiration, and relax the cricopharyngeus muscle to allow passage of the food bolus. The esophageal phase then propels the food bolus distally by means of sequential peristaltic contractions. Alteration of the timing or execution of any of these phases can result in dysphagia, or difficulty swallowing.

Urgencies and Emergencies

A. Acute angioedema and Ludwig angina

Acute angioedema is characterized by localized swelling of subcutaneous and submucosal tissue of the head and neck. The swelling may begin with mild facial involvement but may progress to involve the oral cavity, tongue, pharynx, and larynx. It is often self-limited but may present as a medical emergency. Tongue or laryngeal involvement may rapidly lead to airway obstruction and asphyxiation. Angioedema has a rapid onset and with proper medical treatment usually resolves within 24-48 hours. The underlying pathophysiology of angioedema involves vasoactive mediators such as bradykinin and histamine, causing interstitial edema through endothelial-mediated vasodilation of arterioles with subsequent capillary and venule leakage. The three major etiologies are drug-induced, hereditary, and allergic angioedema. Drug-induced and hereditary angioedema appear to be mediated by the kallikrein-kinin system while allergic angioedema appears to be mast cell-mediated.

Initial management of acute angioedema is focused on airway maintenance. Depending on clinical presentation, this may include nebulized epinephrine inhalation, intubation (either oral or transnasal fiberoptic), or tracheotomy. Due to the potentially rapid progression of this disease process, there should be a low threshold for securing an airway by the latter two methods. Many authors advocate the use of glucocorticoids (10 mg dexamethasone IV every 8 hours) combined with histamine receptor antagonists (both an H1 antagonist such as benadryl 25 mg IV every 6 hours, and an H2 antagonist such as ranitidine 50 mg IV every 6 hours) for 24 hours. Recent evidence has shown that the latter therapies may not be beneficial in drug-induced or hereditary angioedema. This should be balanced with the understanding that in clinical practice, rapid differentiation of the exact subtype of angioedema may not be as practical as empiric therapy given the relatively benign nature of these medications compared to the potentially life-threatening threat of airway obstruction.

Drug-induced angioedema has classically been associated with the use of angiotensin-converting enzyme inhibitors (ACE inhibitors), although many other medications can also (less commonly) cause this phenomenon. The incidence of angioedema secondary to ACE inhibitor use has been noted to be 0.4%-0.7%. The pathophysiology of ACE inhibitor-induced angioedema appears to be secondary to a localized increase in bradykinin levels related to the inhibition of ACE. Studies have shown that one-half of all cases of ACE inhibitor-induced angioedema occur within the first week of initiating treatment. However, some patients undergo years of ACE inhibitors therapy without incident before their first attack of acute angioedema. Treatment should begin with emergent airway maintenance with discontinuation of any possible drugs that may be inducing the edema. Other drugs with known complications of angioedema include rituximab, alteplase, fluoxetine, laronidase, lepirudin, and tacrolimus. Studies involving angiotensin II receptor antagonists (ARBs) have revealed a decreased incidence of angioedema in comparison to ACE inhibitors therapy. However, clinicians should still use caution when initiating an ARB in a patient with known ACE inhibitor-induced angioedema.

Hereditary angioedema involves a deficiency or dysfunction of C1-esterase inhibitor, which leads to increased levels of vasoactive bradykinin. It is autosomally dominant in inheritance and the defect has been mapped to chromosome 11q. Clinically, hereditary angioedema often presents with recurrent episodes of facial and oral swelling, as well as abdominal pain secondary to intestinal wall edema. Studies have revealed that certain drugs (estrogen, ACE inhibitors, ARBs), surgery, and infections may elicit acute angioedema attacks in hereditary angioedema patients. After airway maintenance has been ensured, intravenous C1-esterase inhibitor is the treatment modality of choice. The synthetic steroid danazol has also been used prophylactically to help prevent future acute episodes by increasing the functional levels of C1-esterase inhibitor.

Allergic angioedema is mast cell-mediated and histamine plays the major role in its pathophysiology. In contrast to drug-induced and hereditary angioedema, skin changes including urticaria and pruritis are commonly seen in allergic angioedema. Clinically, the pruritic wheals are spread by scratching and lesions are usually limited to the lips and periorbital areas and less commonly the extremities and genitalia. This form of angioedema is frequently seen in patients who also suffer from atopic dermatitis, allergic rhinitis, and asthma. Triggers of acute attacks of allergic angioedema include certain drugs, infections, food, and plant products.

Ludwig angina is an uncommon life-threatening condition characterized by cellulitis involving the submental, sublingual, and submandibular spaces. The source of the infection is odontogenic and spreads rapidly. The infection is usually polymicrobial with aerobic and anaerobic gram-positive cocci and gram-negative rods. Prior to antibiotics the mortality rate exceeded 50%. Clinically, patients present with painful neck swelling and edema of the floor of mouth often leading to elevation and displacement of the tongue. Patients will have a prominent “hot potato” voice and palpation of the floor of mouth reveals woody edema. The most common cause of death in these patients is airway compromise, and therefore, primary treatment should be centered on airway maintenance with early involvement of anesthesiology and otolaryngology physicians. Intubation is often anatomically difficult and tracheotomy under local anesthesia is often the preferred method of ensuring a patent airway. Recent case studies have shown the benefit of intravenous dexamethasone and nebulized epinephrine to aid with temporizing the airway and transnasal intubation. Airway observation in the intensive care unit may be an option for less severe cases. After the airway has been secured, proper systemic antibiotic therapy should be initiated immediately followed by incision and drainage for most cases. Major complications include extension of the infection posteriorly to involve the parapharyngeal and retropharyngeal spaces as well as the superior mediastinum. The diagnosis is clinical, but CT may be used to assess the retropharyngeal extension of the infection.

B. Peritonsillar Abscess

Peritonsillar abscess (PTA) is a common infection of the peritonsillar space between the palatine tonsil, tonsillar pillars, and the superior pharyngeal constrictor muscle. Its prevalence in the United Stated has been estimated at 30 per 100,000 people annually. The infection is suppurative in nature and thought to be secondary to either extension of adjacent acute tonsillitis or obstruction of Weber glands (minor salivary glands) at the tonsillar pole. PTA is seen in both children and adults. The typical patient presents complaining of a 4- to 5-day history of sore throat and fever, with worsening trismus, odynophagia, dysphagia, and inability to tolerate secretions. PTA should be managed acutely, as the infection may progress and spread to the deep neck tissue and compromise the airway.

The gold standard in diagnosis of PTA is physical examination, which reveals bulging soft tissue, tonsillar erythema and exudate, and possible uvular deviation. Needle aspiration or incision and drainage confirms the diagnosis. Recent studies have revealed the benefit of intraoral ultrasound in the diagnosis of PTA. CT may be necessary in patients with severe trismus or young uncooperative patients and can aid in differentiating PTA from retropharyngeal abscess.

The management of PTA is dependent on the patient characteristics. Needle aspiration can be performed quickly, is relatively safe, and can be both diagnostic and therapeutic. Incision and drainage should only be performed by physicians with an understanding of the relevant pharyngeal anatomy, as multiple vital structures including cranial nerves and the carotid artery potentially lie within the surgical field. In cooperative adults, it can usually be performed using local anesthesia but patients with severe trismus may require general anesthesia. In our practice (for adult patients), preprocedure administration of 900 mg clindamycin, 10 mg dexamethasone, IV fluid hydration, and IV morphine for analgesia can greatly facilitate the procedure by reducing trismus and promoting patient comfort and cooperation. After instillation of local anesthetic, a needle aspiration can be used to confirm the site of incision. A limited mucosal incision is made with a scalpel, taking care not to penetrate the underlying muscular layer. Blunt dissection is then used to penetrate the abscess cavity. Oral suction should be in place before penetration of the abscess cavity to prevent possible aspiration of purulence. Studies have shown both needle aspiration and incision and drainage to be greater than 90% effective but both carry a 10%-15% risk of recurrent PTA. The pediatric patient usually will not tolerate either needle aspiration or incision and drainage under local anesthesia and general anesthesia is often necessary. Antibiotic therapy against Streptococcus pyogenes and oral anaerobes with either penicillin or clindamycin should follow abscess drainage. Most patients with PTA can be treated as outpatients.

Emergent tonsillectomy in the setting of acute infection (“Quincy tonsillectomy”) is usually reserved for cases of PTA that are not successfully drained by simple incision and drainage. While some authors advocate this as a first-line therapeutic option, the increased operative difficulty secondary to acute inflammation, and increased incidence of postoperative hemorrhage make this a less attractive option. It requires general anesthesia and most otolaryngologists would advocate overnight hospitalization postoperatively for observation.

If needle aspiration or incision and drainage of a PTA is performed, it is important to counsel the patient that future risk of PTA is increased. Many otolaryngologists therefore recommend an elective tonsillectomy be performed 2-3 months following a PTA.

C. Deep Neck Space Infections

Although greatly decreased in incidence since the advent of modern antibiotics, deep neck space infections remain a potentially life-threatening condition and require acute recognition and treatment. The source of deep neck space infection is most commonly odontogenic. Other sources of infection include adjacent tonsillar, upper respiratory and salivary infections as well as instrumentation and foreign bodies. These infections are usually polymicrobial, composed predominantly of anaerobes mixed with various streptococcal and staphylococcal species. Drug-resistant bacteria are commonly seen in intravenous drug users who present with deep neck space infections secondary to needle injection disruption of the cervical fascia. Presenting symptoms depend on the exact location of the abscess, but patients commonly present with fever, painful sore throat, decreased neck range of motion, dysphagia, and odynophagia. Physical examination can reveal trismus, a toxic appearance with facial and neck edema, cervical lymphadenopathy, and purulent oral secretions. It is not uncommon for patients to present several days after being discharged on antibiotics for a less severe, local infection. It is important to note that patients on antibiotic or immunosuppression therapy may have more subtle signs of infection and systemic toxicity may be masked.

A thorough understanding of the anatomy of the cervical fascia and deep spaces of neck is paramount in diagnosing and managing these rapidly spreading infections. The cervical fascia is divided into two layers, the superficial cervical fascia and the deep cervical fascia. The deep cervical fascia is further divided into three separate layers—superficial layer of deep cervical fascia, middle layer of deep cervical fascia, and deep layer of deep cervical fascia. Furthermore, the middle layer of deep cervical fascia is subdivided into muscular and visceral divisions and the deep layer of deep cervical fascia is subdivided into alar and prevertebral layers.

These divisions of the deep cervical fascia separate the neck into numerous potential spaces, which may harbor these life-threatening infections. One easy way of categorizing these spaces is by their relationship to the hyoid bone. Potential spaces that exist entirely above the hyoid bone include the submandibular space and the parapharyngeal space. The pretracheal space exists entirely below the hyoid bone. The prevertebral space, danger space, and retropharyngeal space extend along the entire length of the neck.

Parapharyngeal abscess must be differentiated from PTA, as the former requires external drainage via the submaxillary fossa while the latter is best drained intraorally. Retropharyngeal abscesses are most commonly seen in children and are located between the visceral layer of the middle layer of deep cervical fascia and the alar division of the deep layer of deep cervical fascia. Because of the smaller caliber airway in children, retropharyngeal abscess represents a potential source of airway obstruction and should be managed accordingly. The danger space is the region between the alar and prevertebral layers of the deep cervical fascia, which extends from the base of the skull to the diaphragm. Infection of the danger space usually arises from contiguous spread from retropharyngeal, prevertebral, or pharyngomaxillary infections. The lack of definitive anatomic barriers in the danger space offers very little resistance to spread of infection. The prevertebral space is directly posterior to the danger space between the prevertebral division of deep cervical fascia and the paravertebral fascia. Infection of prevertebral space is usually secondary to penetrating trauma or tuberculosis.

Lateral and anteroposterior neck films have traditionally been used to help localize an abscess, although this has now largely been replaced by CT. Recent studies have shown a potential benefit of MRI in delineating soft tissue and vascular spread of infection, although the time required to obtain MRI imaging makes its use of questionable value. Deep neck space infections can progress rapidly and patients usually require hospitalization under close supervision.

Initial treatment should focuses on an evaluation of need for securing an airway with either endotracheal intubation or tracheotomy. Once airway patency has been verified or secured, needle aspiration can sometimes be performed (in the case of easily accessible abscesses) to obtain culture and gram stain. Intravenous antibiotics are started immediately and should cover both aerobic and anaerobic organisms. Ampicillin-sulbactum or clindamycin are commonly used. If methicillin-resistant Staphylococcus aureus is suspected, vancomycin should also be initiated. If the patient does not improve clinically on antibiotic therapy after 48 hours, open surgical drainage of the abscess may be necessary. Major complications of deep neck space infections include mediastinitis, osteomyelitis, Horner syndrome, and cranial nerve deficits. Involvement of the carotid sheath may also lead to suppurative jugular thrombophlebitis (Lemierre syndrome). Early diagnosis and proper treatment may help limit these serious complications.

Disorders and Diseases

A. Oral Cavity Lesions

Inspection and palpation of the oral cavity is an essential part of the head and neck examination. There are numerous lesions that may affect the oral cavity and this review focuses on those which may predispose patients to oral cavity neoplasms. Approximately half of all head and neck cancers occur in the oral cavity.

Leukoplakia is actually a descriptive term rather than a true pathologic term. It represents an asymptomatic white plaque that cannot be scraped off. It is frequently found on the oral and buccal mucosa as well as the tongue. Its prevalence has been estimated from 1% to 5% of the population. Some studies have linked leukoplakia to tobacco use although the true etiology is still uncertain. Leukoplakia represents the clinically evident result of hyperplastic epithelial growth. Many authors recommend a short 1-2 week trial of oral topical steroid preparation (kenalog in orobase) for initial management. Leukoplakia is associated with a malignant transformation rate of approximately 5%, and persistent lesions should be biopsied to assess for premalignant dysplasia or malignancy. The rates of dysplasia are highest for lesions found on the floor of the mouth, tongue, and lower lip. Treatment is aimed at preventing malignant transformation to oral cavity squamous cell carcinoma (OCSCC). Surgical excision, KTP laser, and CO₂ laser excision have all been shown to be effective. Recent studies have also revealed potential efficacy of medical treatment with both topical bleomycin in dimethyl sulfoxide and retinoid compounds.

Erythroplakia is categorized as a nonhomogenous leukoplakia and is described as a velvety red plaque that cannot be removed. It is found in similar regions of the oral cavity as leukoplakia and is also usually asymptomatic at presentation. It is less common than traditional homogenous white leukoplakia with prevalence estimated at 0.2%-0.8%. However, it boasts a higher degree of premalignant dysplasia than leukoplakia, with over half of cases having in situ or overtly invasive squamous cell carcinoma on histologic examination.

Lichen planus is a common dermatologic lesion that may present in the oral cavity. The skin lesions are classically described as pruritic, planar, purple, polygonal papules, while the oral lesions have several different phenotypic subtypes. Reticular, plaque-like, atrophic, erosive, and bullous forms have all been described in the literature. The oral lesions have a questionable capacity for malignant transformation. The data is controversial but one study has estimated the rate of malignant transformation to OCSCC between 1% and 5%. The lesions require biopsy to confirm the diagnosis. Treatment includes vigorous oral hygiene, topical steroids, and immunosuppressive agents.

B. Sialadenitis and Sialothithiasis

There are several nonneoplastic and inflammatory conditions that may affect the major and minor salivary glands. Infection of the salivary glands may be either viral or bacterial in etiology. Viral sialadenitis is most commonly secondary to mumps, which presents with a flu-like prodrome followed by parotid gland swelling. Mumps normally affects children and may be complicated by orchitis, oophoritis, aseptic meningitis, and encephalitis. The incidence of mumps has declined significantly since routine vaccination has been instituted. Other viruses known to cause sialadenitis include cytomegalovirus, coxsackievirus A and B, echovirus, Epstein-Barr virus, and influenza A.

Bacterial sialadenitis may be acute or chronic. Acute suppurative sialadenitis occurs most commonly in dehydrated patients who are postoperative, elderly, or those on diuretic therapy. The parotid gland is affected in the majority of cases secondary to the diminished bacteriostatic activity of its serous saliva. Salivary stasis, ductal obstruction, and decreased saliva production appear to be predisposing conditions to acute sialadenitis. Patients usually present with fever, systemic toxicity, and tender swelling and enlargement of the affected glands. The most common organism isolated is S aureus, though culture may reveal a polymicrobial infection with both aerobic and anaerobic organisms. Treatment involves ample hydration, warm facial compresses, sialogogues (such as lemon wedges) to stimulate saliva secretion in the affected gland. Antibiotics targeted against S aureus should be started immediately and continued for 7-10 days. CT may be necessary to rule out abscess formation or stone (sialolithiasis) in patients who do not improve clinically after several days of appropriate therapy.

Sialolithiasis may occur in the setting of acute or chronic sialadenitis or it may be an incidental finding on routine imaging studies. Salivary calculi affect males more than females and are seen most frequently between ages 30 and 60. In contrast to sialadenitis, sialolithiasis preferentially affects the submandibular glands because of their alkaline, high calcium, mucus-rich environment. Large, solitary, radio-opaque stones are usually found in the submandibular glands, while the parotid glands are more likely to have multiple smaller, radiolucent stones. Calculus formation is believed to be secondary to partial obstruction of the salivary duct combined with calcium-rich stagnant saliva. Salts composed of calcium phosphate, magnesium, ammonium, and carbonate precipitate in this environment. Contributing factors to the development of salivary calculi include underlying acute or chronic sialadenitis, dehydration, and anticholinergic medications. Uric acid salivary calculi may also be seen in the setting of gout.

Patients with sialolithiasis frequently present with pain and swelling of the affected gland, although many patients have asymptomatic calculi discovered incidentally. Eating usually exacerbates the pain. Physical examination may reveal the location of the calculi by simple palpation. Obstruction of the flow of saliva can be analyzed by massaging the gland. It is important to note that stones are more commonly found in the salivary ductal structures rather than their associated glands. CT is the preferred method of imaging if sialolithiasis is suspected but a calculus is unable to be palpated on physical examination. CT has a 10-fold greater sensitivity than plain films in detecting salivary calculi. Ultrasound may also have benefit in locating calculi when CT is unavailable. Sialography is no longer routinely used and is contraindicated in patients with acute sialadenitis.

Treatment of sialolithiasis should begin conservatively with hydration, salivary gland massage, heating pads applied to the affected gland, and sialogogues. Anticholinergic medications should be discontinued and antibiotics should be initiated if there is concern for acute suppurative sialadenitis. There are several options for more invasive therapy for patients who do not respond to conservative management. Transoral removal of submandibular stones, sialadenectomy, lithotripsy, wire-basket removal, and sialoendoscopy have all been shown to be effective in the appropriate patient.

C. Acute and Chronic (Recurrent) Tonsillitis

Tonsillitis is one of the most common problems encountered by the otolaryngologist. In general, tonsillitis refers to inflammation of the palatine tonsils located on the lateral walls of the oropharynx between the palatoglossal and palatopharyngeal folds. The pharyngeal tonsils or adenoids, while part of Waldeyer’s ring of lymphatic tissue are anatomically separate from the palatine tonsils. The adenoids are located on the posterior wall of the nasopharynx in close proximity to the eustachian tube opening.

Acute tonsillitis is primarily a pediatric disease that tends to affect children aged 5-15. It is most commonly caused by group A beta-hemolytic streptococcal species, although anaerobes, Haemophilus influenzae, and viruses can also be causative agents. Patients usually present with fever, painful sore throat, halitosis, and dysphagia. It is important to note the distinction in clinical presentation between acute tonsillitis, viral pharyngitis, and infectious mononucleosis (caused by the Epstein-Barr virus). Viral pharyngitis commonly presents with a triad of cough, coryza, and conjunctivitis, while patients with mononucleosis typically present with anterior and posterior cervical lymphadenopathy, odynophagia, and a grayish tonsillar exudate. On physical examination, the acute tonsillitis patient has erythematous tonsils, purulent tonsillar exudate, and anterior cervical lymphadenopathy.

Tonsillar hyperplasia is measured in the medial-to-lateral plane of the oropharynx and helps in the assessment of upper airway obstruction. They are traditionally rated on a 1–4 scale, with 1+ being confined below the level of the tonsillar pillars, 2+ at the pillars, 3+ extending past the pillars, and 4+ meeting in the midline. Tonsillar hypertrophy in the pediatric population may predispose to the development of sleep-disordered breathing. Patients typically present with heroic snoring, voice changes, observed episodes of sleep apnea, and daytime somnolence. In these cases, elective tonsillectomy (usually combined with adenoidectomy) can often provide resolution of obstructive symptoms.

The diagnosis of acute tonsillitis is clinical although many practitioners rely on a positive streptococcal test or culture. The mainstay in treatment of acute tonsillitis continues to be a 7-10 day course of penicillin or a comparable cephalosporin. However, it is now estimated that failure with penicillin therapy occurs in up to 30% of cases. Following up culture results and adjusting the antibiotic therapy has therefore become an important step in management. Complications from acute tonsillitis include the formation of PTA and neck abscess formation. Antibiotic use has greatly decreased the incidence of rare systemic complications like poststreptococcal glomerulonephritis and rheumatic fever.

Many patients have a solitary episode of acute tonsillitis, which responds favorably to antibiotic treatment, while some patients have recurrent acute tonsillitis with multiple infections over the course of years. Chronic tonsillitis is a state of persistent tonsillar inflammation for greater than 3 months following an episode of acute infection. It is characterized by chronic sore throat and odynophagia with tonsillar enlargement, tonsillolithic debris, and cervical lymphadenitis present on physical examination. Recent studies have suggested that polymicrobial infections, drug-resistant organisms, H influenzae, S aureus, anaerobes, and actinomycetes may all play a role in chronic tonsillar disease. Treatment of chronic tonsillitis, therefore, should begin with the use of broader spectrum antibiotics like amoxicillin-clavulanate or clindamycin, which may target these offending organisms.

The critical question in the treatment of both acute and chronic tonsillitis is whether or not to perform a tonsillectomy. Medical management with antibiotics is always the first-line therapy for acute tonsillitis. Tonsillectomy for acute episodes (also known as “Quincy tonsillectomy”) is usually considered only when complications such as deep neck-space abscess or acute airway obstruction are concurrently present. Tonsillectomy is generally performed for patients with recurrent acute tonsillitis, defined as seven episodes within 1 year, five episodes each year for the past 2 years, or three episodes per year for 3 successive years.

D. Obstructive Sleep Apnea

Obstructive sleep apnea is an intrinsic dyssomnia that affects roughly 15-20 million people in the United States. It is classically described as the presence of hypopneic episodes, apneic episodes, and respiratory effort related arousals that occur during sleep. An apneic event requires 10 or more seconds of cessation of airflow, followed by arousal with restoration of normal ventilation. The definition of hypopnea varies between sleep laboratories, but generally refers to an episode of decreased airflow (>50% reduction) for greater than 10 seconds, associated with decreased oxygen saturation or arousal. Patients usually present with nighttime snoring, daytime hypersomnolence, irritability, morning headaches, cognitive impairment, and often witnessed apneic events with cessation of airflow followed by choking or gasping. When taking a history, it is important to confer with the patient’s bed partner, as the patient may be unaware of his/her nighttime symptoms. Cardiovascular disease, hypertension, metabolic dysfunction, respiratory failure, and cor pulmonale are among the serious long-term effects of obstructive sleep apnea.

The pathophysiology of obstructive sleep apnea appears to be a combination of upper airway collapse and decreased neural output from the respiratory center in the brainstem. During normal inspiration, the pharyngeal muscles are stimulated via a central nervous system reflex pathway to help maintain pharyngeal airway patency. During sleep, however, these neural reflexes are attenuated and the airway becomes more susceptible to collapse. Patients who are predisposed to airway obstruction due to anatomic reasons are at high risk for developing obstructive sleep apnea. Documented anatomic risk factors for obstructive sleep apnea include macroglossia, adenotonsillar hypertrophy, elongation of the soft palate, and retrognathia.

Obstructive sleep apnea is more common in men and the overall incidence appears to increase with age. The most significant risk factor is obesity and the recent increase in prevalence is thought to relate to the current obesity epidemic. Other known risk factors include nasal obstruction, smoking, diabetes mellitus, alcohol consumption, and the previously mentioned anatomic abnormalities. Physical examination may reveal obesity with an increased neck circumference. A thorough head and neck examination should be performed to fully assess airway patency. Oral cavity examination may reveal tonsillar hyperplasia. The nasal cavity and nasopharynx should be examined using a flexible fiberoptic endoscope to rule out nasal obstruction secondary to deviated septum, nasal polyps, or turbinate hyperplasia. The modified Mü ller maneuver should be performed to assess for site of upper airway collapse during inspiration. Patients are asked to inspire against a closed mouth while their nose is pinched shut, thus creating a column of negative pressure within the upper airway. The airway at the level of the soft palate, lateral pharyngeal wall, and base of tongue are then observed for luminal collapse and graded on a 1-3 scale.

Polysomnography remains the gold standard in diagnosing obstructive sleep apnea, which can be defined by either the apnea-hypopnea index (AHI) or the respiratory disturbance index (RDI). The AHI includes the number of hypopneas and apneas which occur per hour of sleep, while the RDI is the number of hypopneas, apneas, and respiratory effort related arousals per hour of sleep. The generally accepted guidelines for diagnosis of obstructive sleep apnea are an AHI of 15 or greater in an asymptomatic patient or an AHI of greater than 5 in a symptomatic patient. Imaging studies are usually not required to diagnose obstructive sleep apnea, although they may help assess for upper airway anatomic abnormalities.

Treatment of obstructive sleep apnea should begin with the identification and prevention of risk factors. Behavior modifications such as weight loss, smoking and alcohol cessation, and discontinuation of any CNS depressants may help improve the patient’s apneic index. Safety precautions related to daytime hypersomnolence need to be discussed for high-risk patients like pilots and commercial truck drivers.

Continuous positive airway pressure (CPAP) is considered first-line therapy after behavioral modifications. The positive pressure helps keep the airway patent and has been shown to be highly effective in reducing obstructive sleep apnea symptoms. Some patients experience difficulty tolerating CPAP therapy. Breathing against the positive pressure airflow can be difficult to adjust to, and this must be considered when treating the obstructive sleep apnea patient. Bi-level positive airway pressure (BiPAP) is another system that employs a higher inspiratory pressure with a lower expiratory pressure to allow for easier expiration. There are numerous oral appliances, which may help adjust the airway during sleep to improve patency, although studies have shown these appliances to be less effective than CPAP in improving the apneic index.

Several surgical options are available for patients who do not respond well to initial treatment. Routine adenoidectomy and tonsillectomy has been shown to be effective in pediatric patients with isolated adenotonsillar hypertrophy contributing to their obstructive sleep apnea symptoms. For adult patients, more extensive surgery is usually required. Uvulopalatopharyngoplasty (UP3) is the first-line surgical therapy in which the uvula, a small amount of soft palate, and the palatine tonsils are removed. The concept of “multi-level sleep surgery” is important, as UP3 alone is associated with a significant (>50%) failure rate. Most commonly UP3 is combined with additional procedures to address other areas of obstruction. Septoplasty and inferior turbinate reduction can improve nasal airflow. The tongue base can be advanced by suture suspension, hyoid suspension, or genioglossus advancement. Other more invasive surgical interventions include maxillomandibular advancement. Some authors have advocated less-invasive therapy including palatal stiffening implants and radiofrequency ablation, but the long-term data are lacking. Life-threatening obstructive sleep apnea that does not respond to CPAP/BiPAP or surgical intervention may require permanent tracheostomy. A repeat polysomnogram should be performed in surgical patients 1-3 months postoperatively to monitor for improvements.

Anatomy and Physiology

The anatomy of the larynx (Figures 15–7 and 15–8) is best understood in the context of its function. The primary function of the larynx is to protect the airway from aspiration during swallowing. The epiglottis and aryepiglottic folds help direct food and liquids laterally into the pyriform sinuses and away from the midline laryngeal inlet. The paired arytenoid cartilages act as attachment points for most of the intrinsic muscles of the larynx, serving to move the vocal folds together (adduction) and apart (abduction). The false vocal folds and true vocal folds adduct to prevent entry of food or liquids into the airway.

The larynx also functions in respiration. Owing to reflex pathways in the brainstem, the glottis opens just prior to inspiration. Other laryngeal reflexes respond to subglottic pressure and hypercapnea. Additionally, initiation of swallowing causes a reflex period of involuntary apnea.

Phonation is the most uniquely human of the three functions of the larynx. At its most basic description, the glottic larynx produces a fundamental tone by the vibration of the free edge of the true vocal folds. This vibration is due to passive vibration of the vocal folds from air moving past the opposed free edges. Changing the tension within the vocal fold changes the pitch at which the vocal fold vibrates.

Several cartilages comprise the framework of the larynx (Figure 15–8). The thyroid cartilage is the largest laryngeal cartilage. This shield-shaped cartilage is responsible for the anterior neck prominence, sometimes called the “Adam’s apple” in lay terminology, and it provides protection to the internal components of the larynx. The cricoid cartilage lies inferior to the thyroid cartilage and serves as the major support for the larynx. It is the only complete cartilaginous ring within the upper airway. Internally, the paired arytenoid cartilages articulate with the cricoid cartilage, and attach to the vocal folds. Movement of the arytenoids results in abduction or adduction of the vocal folds. The epiglottis is a flexible cartilage located above the larynx. It is not involved in structural support of the larynx, but serves to assist in protecting the airway during deglutition (swallowing).

Innervation to the larynx is provided by the vagus nerve (cranial nerve X). The vagus nerve originates from three nuclei located within the medulla—the nucleus ambiguous, the dorsal nucleus, and the solitary tract nucleus. All motor fibers (and thus laryngeal motor innervation) originate from the nucleus ambiguous. The dorsal (parasympathetic) nucleus is the origination for efferents to involuntary muscles of the bronchi, esophagus, heart, stomach, and intestine. Sensory innervation from the pharynx, larynx, and esophagus terminates in the solitary tract nucleus. The vagus nerve exits the skull base through the jugular foramen. It descends in the neck behind the jugular vein and carotid artery and sends pharyngeal branches to the muscles of the pharynx and soft palate. The superior laryngeal nerve arises directly from the vagus, and has an internal and external branch. The internal superior laryngeal nerve enters the larynx through the thyrohyoid membrane and supplies sensation to the larynx above the true vocal cords. The external superior laryngeal nerve innervates the cricothyroid muscle, the only muscle of the larynx not innervated by the recurrent laryngeal nerve. The right recurrent laryngeal nerve arises from the vagus and loops around the subclavian artery. The left recurrent laryngeal nerve arises more distally in the thorax, and loops around the aortic arch. Both recurrent laryngeal nerves then ascend in the tracheo-esophageal grooves, and enter the larynx near the cricothyroid joint. The recurrent laryngeal nerves provide motor innervation to all of the intrinsic muscles of the larynx except the cricothyroid. A summary of the laryngeal muscles is provided in Table 15–4.

Urgencies and Emergencies

A. Pediatric Airway Obstruction

The rapid and accurate evaluation of a child in respiratory distress is one of the most critical skills an otolaryngologist must master. Noisy breathing and respiratory distress have multiple etiologies, and differentiation between acute emergencies and chronic conditions is essential. An important point to learn early on is that not all noisy breathing is stridor.

True stridor may represent an impending airway failure, and thus must be distinguished from other upper airway noises that are sometimes mistakenly referred to as “stridor.” Airway obstruction at the level of the nasopharynx produces snoring sounds, or stertor. Tracheobronchitis can produce a wheezy, barking cough characteristic of croup. Asthma, tracheobronchial foreign bodies, and bronchomalacia can produce wheezing. “Stridor” specifically refers to the noise that is produced by air movement through a partially obstructed airway. Inspiratory stridor usually signifies an obstruction above the level of the vocal cords, while expiratory stridor most often occurs with subglottic obstruction. Biphasic stridor usually signifies an obstruction at the level of the vocal cords or subglottis.

The immediate concern in evaluating a child with stridor is verifying or establishing a stable airway. The initial evaluation should consist of non-invasive examination to avoid exacerbating a potentially unstable airway. If the child is in acute respiratory distress, evaluations such as flexible fiberoptic laryngoscopy should not routinely be performed.

A careful history should be obtained, including the duration of the stridor and relationship to feedings. Parents should be questioned about any change in symptoms with position changes. Presence of any birth or intrauterine complications, history of intubation, as well as congenital anomalies should be determined.

If the child is not in acute distress, flexible fiberoptic laryngoscopy may provide valuable diagnostic information. This can be done with the child awake and restrained, or under general anesthesia with spontaneous ventilations. This allows determination of nasopharyngeal and supraglottic anatomy, as well as vocal cord motion. If available, the examination should be recorded to allow playback and review, as real-time examination can be problematic in an often-uncooperative child.

Other investigations should be based on the presence of suggestive symptoms. Swallowing function is often impaired in children with stridor, and should be evaluated. Vascular rings may produce extrinsic compression of the esophagus and trachea leading to stridor, feeding difficulties and failure to thrive. An altered, weak, or absent cry from birth can suggest neurologic impairment. Recurrent pneumonia or excessive cough with feeding may be present with vocal cord impairment, severe reflux, or tracheoesophageal fistula.

One of the most common causes of pediatric stridor is laryngotracheobronchitis, or croup. This is an acute viral illness, most commonly caused by the parainfluenza virus. The typical patient is an infant or young child with low-grade fever, seal-like barking cough, and occasionally biphasic stridor. The classic radiographic finding is a “steeple” sign visible on AP views, indicative of the characteristic narrowed subglottic airway from edema. The typical course for most patients is resolution over several days and few patients require hospitalization. Signs of respiratory distress including tachypnea, retractions, and cyanosis may necessitate closer observation such as hospitalization. For these more severe cases, treatment with humidified air, nebulized racemic epinephrine, and systemic steroids may be indicated. Manipulation of the airway may exacerbate the clinical situation, and should be avoided unless clearly indicated.

Epiglottitis is fortunately becoming vanishingly rare in most industrialized countries, owing to nearly universal vaccination of children against H influenzae type B. If recognized and managed appropriately (with aggressive airway control) outcomes are usually excellent. If managed conservatively, epiglottitis is associated with up to 6%-10% mortality. Patients present with high fever, drooling, and odynophagia and are usually toxic in appearance. Of note, epiglottitis tends to progress quite rapidly; patients can decompensate clinically in a matter of hours. The characteristic position naturally assumed by patients with epiglottitis is the leaning forward position, to maximize their marginal airway opening. Even oral cavity examination with a tongue blade can precipitate an airway crisis; therefore evaluation and treatment ideally consists of immediate control of the airway in the operating room under general inhalational anesthesia. A cherry-red epiglottis will be visible on endoscopic examination. Pharyngeal and blood cultures should be obtained, and the child started on broad spectrum IV antibiotics such as ceftriaxone. Once definitive identification of the culprit organism is made antibiotic selection can be narrowed appropriately. Children are left intubated until air leak around the endotracheal tube is evident.

Chronic pediatric stridor is most often due to laryngomalacia. Parents will usually report onset of symptoms shortly after birth. It often worsens initially, but in the vast majority of cases resolves without the need for intervention, usually by 12-18 months. A variety of factors have been hypothesized to contribute to laryngomalacia including neurologic, muscular, and reflux-induced inflammation. The stridor is worsened while crying or in an excited state, and is usually relieved by placement in the prone position. Flexible fiberoptic laryngoscopy reveals collapse of floppy supraglottic structures such as the epiglottis and aryepiglottic folds. There is a strong association of laryngomalacia with reflux, and presumptive treatment with an acid blocking medication (such as ranitidine) may be beneficial. Surgical treatment of the supraglottis is reserved for severe cases such as patients with cyanosis or failure to thrive. The most common procedure is aryepiglottiplasty, where cold knife or carbon dioxide laser is used to excise redundant mucosa over the arytenoid cartilages. Rarely, tracheostomy may be necessary.

Subglottic stenosis is the second most common cause of chronic pediatric stridor. Causes include both congenital stenosis, and acquired (most often from prolonged intubation). Typically, patients have recurrent “croup” and biphasic stridor. If present in older children, the patient may only become symptomatic periodically, in association with upper respiratory tract infections. Diagnosis requires endoscopic evaluation, and can be defined as a subglottic airway diameter of less than 4 mm in full-term infants and less than 3 mm in premature infants. Pediatric subglottic stenosis is traditionally graded according to the Cotton scale (Table 15–5). Tracheotomy may be required in moderate to severe cases. Treatment of subglottic stenosis can consist of tracheal dilatation (either by rigid serial dilation or controlled radial expansion balloon), debridement by microdebrider or carbon dioxide laser, cricoid split, or laryngotracheoplasty.

B. Foreign Body Aspiration

Airway foreign body typically involves children between 1 and 4 years of age, but can occur in any age group. Children in this age range tend to place objects in their mouths and lack molars for grinding of food. Objects can range from peanuts (most common) to coins, marbles, and toy products. Adult foreign body aspiration is usually associated with food, typically meat. Foreign body aspiration is potentially life threatening and it is recognized as the fifth most common cause of unintentional-injury related mortality in the United States. Pharyngeal foreign body (most commonly in the vallecula) represents a potential impending airway foreign body, and should be treated as such.

The clinical presentation of patients with airway foreign body depends on the anatomic location. Patients with pharyngeal or hypopharyngeal foreign bodies typically present with dysphagia, odynophagia, and occasionally drooling from inability to tolerate secretions. If large objects become lodged in the larynx, patients may present with pain, dysphonia, inspiratory stridor, and dyspnea. Tracheal foreign bodies produce both inspiratory and expiratory stridor. Distally located foreign bodies (Figure 15–9) often lodge in the right bronchus (especially in adults). This is due to the angles at which the left and right mainstem bronchi branch off the trachea, with the right being a less acute angle. Distal foreign bodies typically produce unilateral wheezing and decreased breath sounds. A history of acute choking episode is very common and is 76%-92% sensitive in diagnosing foreign body aspiration.

Acute management of airway foreign body is dictated by patient condition. In a conscious patient who is able to exchange air and cough, an attempt at foreign body removal should not be made immediately. Unconscious patients, or victims unable to cough or move air require immediate intervention. If the means for performing a cricothyrotomy or tracheotomy are not available, the Heimlich maneuver can be performed as three manual abdominal thrusts to compress the lungs and potentially produce enough airway pressure to dislodge the foreign body. For patients without impending loss of airway, anteroposterior and lateral radiographs of the airway including larynx and chest can be helpful. While only radio-opaque foreign bodies can be visualized, radiographs may demonstrate obstructive emphysema, atelectasis, or consolidation. Additionally, they can provide a baseline study for future comparison.

Definitive removal of an airway foreign body requires general anesthesia, and direct laryngoscopy. Distally located foreign bodies are best addressed with rigid bronchoscopy for removal. A clear history consistent with aspiration often should prompt operative evaluation, even if specific symptoms (stridor, unilateral wheezing, decreased breath sounds) are lacking, as these may be absent in up to 40% of cases. Suspected nut aspiration (common in children) should be treated aggressively. Local tissue reactions to the nut oils and proteins are common, and can be robust. Post removal, patients with peanut aspiration may require intensive care unit observation and ventilatory support while the inflammatory reaction resolves.

C. Laryngeal Trauma

Laryngeal trauma is rare, representing 1 out of every 14,000-42,000 emergency room visits. Quick recognition of laryngeal trauma is essential as it can rapidly lead to death. Up to one-third of laryngeal trauma victims die prior to arrival to a hospital setting.

Laryngeal trauma can be classified as blunt or penetrating. Blunt laryngeal trauma results from crushing of the laryngeal framework against the cervical spine, and usually results from motor vehicle accidents. Other common etiologies include strangulation type injuries, and clothesline injuries. Penetrating laryngeal trauma usually results from projectile injury, such as a gunshot, or knife wounds to the neck.

Every patient with trauma to the neck should be evaluated for potential laryngeal trauma. Patients may report dyspnea, hoarseness, or aphonia. Other less common symptoms include dysphagia, anterior neck pain, and odynophagia. Evaluation should begin with the ABCs of trauma–airway, breathing, circulation. Severe laryngeal trauma (particularly clothesline-type injuries) can result in loss of airway and necessitate the need for immediate tracheotomy. Physical findings common in patients with laryngeal trauma include stridor, subcutaneous crepitance (emphysema), bruising or edema to the anterior neck, loss of palpable landmarks, and hemoptysis. As outlined above (stridor and airway obstruction) the type of stridor can often give an indication of the site of airflow obstruction.

Further evaluation of laryngeal trauma is dictated by the patient’s condition. If an adult patient’s airway is unstable, most experts would advocate awake tracheostomy or cricothyrotomy under local anesthesia, as endotracheal intubation in the setting of laryngeal trauma can be problematic. For unstable pediatric laryngeal trauma, inhalational anesthetic followed by rigid endoscopic intubation is recommended by many experts. Once a stable airway has been determined or secured, evaluation can proceed.

Flexible fiberoptic laryngoscopy is advocated for awake, stable patients. The airway should be evaluated for vocal cord mobility, edema, laryngeal lacerations, and hematomas. Because the underlying mechanism of injury often results in damage to other adjacent structures, laryngeal trauma patients should undergo complete cervical spine radiography, and evaluation for esophageal or vascular injury should be considered. Some authors recommend fine-cut CT imaging of the larynx to help guide treatment planning. It should be emphasized that CT scanning rarely provides useful information regarding the immediate airway management for patients suspected of having laryngeal trauma, but often provide helpful information regarding surgical planning for repair.

Management of the patient with laryngeal trauma depends on the severity of injury. Patients with small laryngeal hematomas, small lacerations not involving the vocal fold edge or anterior commissure, and nondisplaced stable thyroid cartilage fractures can often be managed without tracheostomy. Most such patients should be hospitalized for 24 hours of airway observation and placed on systemic steroids, humidified air, and proton pump inhibitor (PPI) therapy. If mucosal disruption is present antibiotics should be initiated.

More severe laryngeal traumas often require tracheotomy and direct laryngoscopy or even open laryngeal exploration and repair. This should ideally be performed within 24 hours of the initial injury. Open laryngeal exploration is performed via a midline thyrotomy approach. A horizontal skin incision is made at the level of the cricothyroid membrane. Subplatysmal flaps are elevated, and the strap muscles divided at the midline. The airway is then entered in the midline of the cricothyroid membrane, and a vertical incision through the midline thyroid cartilage extended superiorly. Care must be taken to avoid injury to the underlying endolaryngeal mucosa. Following this, the mucosa is incised allowing inspection of the endolarynx. Once the endolarynx is exposed, all mucosal lacerations should be repaired to cover all exposed cartilage. If the anterior commissure is disrupted, a laryngeal stent may need to be placed, although placement of a stent itself leads to some degree of laryngeal injury. If placed, stents should be removed as soon as possible, usually around 2 weeks.

Outcomes for laryngeal trauma patients are fairly good once initial control of the airway is obtained. Most patients usually achieve a stable airway and undergo decannulation, which may take from 1 to 6 months or longer depending on the extent of injury. Overall, up to 90% of patients can recover a satisfactory vocal quality if managed appropriately.

Disorders and Diseases

A. Hoarseness

Hoarseness or dysphonia is defined as an alteration in the quality or character of phonation. Patients may describe their voice as breathy, harsh, or rough. Common etiologies of hoarseness include viral illness, vocal fold paralysis, laryngopharyngeal reflux (LPR), laryngeal polyps, allergy, vocal abuse, dysplasia, and cancer.

Patients should be questioned about onset, frequency, and nature of the hoarseness. As previously discussed, the larynx is an essential part of swallowing, and any history of coughing or choking after eating should be elicited. Likewise the patient should be questioned about recurrent episodes of pneumonia. Any history of intubation, head and neck trauma, or previous head and neck surgery should be sought. Patients should be questioned about smoking and alcohol use.

Physical examination begins with a full head and neck examination. It is important to visualize the larynx. Methods of visualization include indirect mirror examination, rigid endoscopy, or transnasal flexible fiberoptic laryngoscopy. Videostroboscopy offers invaluable information about vocal fold motion and can identify adynamic segments and altered areas of mucosal wave propagation.

One common benign cause of hoarseness is vocal polyps. These are due to local tissue inflammation. Vocal nodules are distinct from polyps, and always occur bilaterally. They are most often a result of vocal abuse/misuse and usually respond dramatically to voice therapy. Vocal fold granulomas are usually due to extra-esophageal acid reflux damage.

Hoarseness can also be the presenting symptom for cancer, most often for cancer of the true vocal folds. The vast majority of laryngeal cancer is squamous cell cancer, and smoking is the biggest risk factor for its development. Laryngeal cancer is discussed further in the section Head and Neck Cancer. While early stage laryngeal cancer is highly curable, advanced stage carries a dramatically reduced prognosis. Thus, any patient with hoarseness lasting longer than 2 weeks should be evaluated and undergo visualization of their larynx.

B. Laryngopharyngeal Reflux

LPR impacts hundreds of thousands of patients annually, with some studies estimating as many as 30% of Americans may suffer from some degree of LPR. It is increasingly clear that LPR is a disease separate and distinct from classic gastroesophageal reflux disease (GERD). Patients with LPR typically present with frequent throat clearing, globus sensation, cough and hoarseness (as opposed to postprandial heartburn with GERD).

Physical examination of patients suspected of LPR should include an examination of the larynx, most commonly by flexible transnasal fiberoptic laryngoscopy. Hoarseness is not pathognomonic for LPR, and can also be present in more serious disorders. Several laryngeal findings are common in patients with LPR. Presence of vocal cord granuloma or pseudosulcus vocalis, although uncommon, is highly suggestive of LPR. Other common laryngoscopic findings consistent with LPR include posterior laryngeal hypertrophy, laryngeal edema and erythema, cobblestoning, or posterior commissure bar.

The diagnosis of LPR relies on a combination of symptoms and physical findings rather than one isolated factor being pathognomonic. Several studies have demonstrated the presence of many of the above symptoms and laryngeal findings in healthy, normal control patients. Many clinicians advocate the use of a metric scale that combines multiple common symptoms or physical findings. Two such instruments are the reflux finding score (RFS) and the reflux symptom index (RSI). In general, an RFS of greater than 8 is suggestive for LPR, and a score of greater than 13 on the RSI is likewise indicative of LPR.

Treatment for LPR currently consists of PPIs as a first-line therapy with surgery (eg, Nissen fundoplication) for selected treatment failures. It should be pointed out that although multiple uncontrolled studies have demonstrated benefit of PPIs for the treatment of LPR, the majority of randomized controlled trials have failed to confirm this. Possible confounding factors include lack of clear “gold standard” for the diagnosis of LPR, and differing treatment regimens. Our current practice is to place patients with LPR on once-daily PPI therapy for a 3- to 6-month period, after which response to treatment is assessed. The extended length of time is critical, as studies have shown that resolution of laryngeal findings can take up to 6 months to resolve once PPI therapy is initiated. If there is no response to once-daily PPI therapy, the patient can be advanced to twice-daily PPI therapy. Disease severity can prompt initiation of PPI therapy at twice-daily dosing; severe laryngeal edema or presence of subglottic stenosis would be two such indications.

It should also be mentioned that if a patient is diagnosed with LPR, some form of evaluation of the esophagus should be undertaken. There is up to a 20% incidence of unsuspected esophageal abnormalities in patients with LPR. This evaluation can take the form of imaging modalities such as barium swallow, or endoscopic examination such as esophagoscopy.

C. Vocal Cord Immobility/Paralysis

Vocal cord mobility problems represent a wide range of etiologies characterized by diverse patient presentation and prognostic outcomes. The distinction between unilateral and bilateral, and between paretic (hypomobile) versus paralyzed cords is imperative.

Patients with unilateral vocal cord paralysis may be asymptomatic, but often present with a hoarse, breathy voice. Their voice often starts out stronger in the morning and worsens throughout the day as they develop vocal fatigue. Accompanying symptoms can include frequent throat clearing, cough, vague globus sensation, and aspiration. Often patients will report subjective shortness of breath or a feeling of “running out of air” despite normal pulmonary function. This is secondary to glottal incompetence (lack of apposition of the vocal folds) resulting in escaped air during phonation. Thus a patient with unilateral vocal fold paralysis may be able to climb a flight of stairs without difficulty, yet feel short of breath when attempting to carry on a telephone conversation. Patients should be questioned specifically about swallowing, weight loss, recent illnesses or intubations, and surgeries (especially cardiac, cervical spine, and thyroid procedures).

The etiologies of vocal cord paralysis reflect the diverse nature of illnesses and injuries that can result in the final common pathway of vocal cord immobility or paralysis. The vocal cords derive their innervation from the vagus nerve, and any injury along the course of this nerve may result in vocal cord paralysis. Unilateral vocal cord paresis is by far the most common, with bilateral representing less than 20% of all paralysis. Historically, the most common cause of unilateral vocal cord paralysis was malignancy (such as lung cancer or skull base tumors). More recent studies show that iatrogenic surgical injury is now the most common cause. Nonthyroid surgical procedures (including anterior approaches to the cervical spine and carotid endarterectomy) now account for the majority of these iatrogenic injuries. Thyroid surgery remains the most common cause of bilateral vocal cord paralysis. Table 15–6 summarizes the most common causes of vocal fold paralysis. It should be noted that laryngeal manifestations of rheumatoid arthritis can rarely mimic vocal cord paralysis, although the underlying problem in this case is vocal cord immobility secondary to fixation of the arytenoid cartilages.

It is important to note that vocal cord immobility or paralysis is a sign of pathology and not a diagnosis. Thus the first concern when evaluating a patient with vocal cord paralysis should be investigation of the etiology. Often the cause is not identifiable, and the vocal cord paralysis is deemed idiopathic. A thorough head and neck examination should be performed, including endoscopic evaluation of the larynx. This most often is by flexible fiberoptic laryngoscopy, with most laryngologists recommending videostroboscopy as well. For unilateral recurrent laryngeal nerve injury, the affected immobile cord will usually lie in a paramedian position. This is due to lack of abduction, with some retained adduction (due to the cricothyroid muscle, which is innervated by the superior laryngeal nerve). For more proximal vagal lesions, the affected cord will usually rest in intermediate position, due to loss of both abduction and adduction innervation. There is also loss of sensation to the affected hemilarynx, and aspiration is common.

Diagnostic testing should include a chest radiograph and CT scan following the entire vagus nerve course (ie, neck and chest, from skull base to the mid-chest). More esoteric tests are likely to be low yield and poorly cost-effective, and should be reserved for more selective use. Many laryngologists advocate the use of laryngeal electromyography (EMG). This test is performed percutaneously, and tests the superior laryngeal nerve and recurrent laryngeal nerve by evaluating motor unit electrical activity in the cricothyroid and thyroarytenoid muscles, respectively. Laryngeal EMG can provide useful information regarding the degree and likely site of injury (central vs peripheral), as well as the potential for spontaneous recovery. It is most predictive if performed 6 weeks to 6 months after initial injury.

Initial therapy for unilateral vocal cord paralysis consists of observation and speech therapy. Often the opposing vocal cord can compensate by crossing the midline and closing the glottal gap. This can produce an acceptable vocal quality, usually occurring within a 3-6 months time span. Patients not obtaining a good result using these conservative measures can be treated by a variety of surgical interventions. The goal of surgical treatment for unilateral vocal cord paralysis is medialization of the affected cord. This reduces the glottic gap and allows the opposing, innervated cord to contact the other vocal fold with less effort. Treatment selection for unilateral vocal cord paralysis depends on the potential for recovery. In cases such as iatrogenic surgical injury with little chance of spontaneous recovery, definitive therapy can be initiated early. For idiopathic causes, a more conservative approach is usually advocated. Overall, up to 60% of patients with idiopathic unilateral vocal cord paralysis will recover to a near-normal voice within 8-12 months. Thus, most experts would recommend waiting at least 1 year before proceeding with definitive therapy. Clear indications for earlier intervention include significant dysphagia and aspiration from glottic incompetence.

Definitive surgical procedures for unilateral vocal fold paralysis include laryngeal framework surgery, injection of longer lasting material, and reinnervation techniques. The thyroplasty technique (a laryngeal framework surgical procedure) is performed through an external skin incision. After exposing the thyroid cartilage, a window is cut in the thyroid ala overlying the position of the vocal fold on the affected side. An implant is then placed in a subperichondrial window, thus pushing the vocal fold toward the midline. Implants can include Silastic, autologous cartilage, and Gore-Tex. Most laryngologists perform this procedure with the patient lightly sedated. A flexible fiberoptic laryngoscope can be suspended in position, and the patient is asked to phonate periodically so that the surgical effects can be evaluated in real time and adjusted accordingly.

Another common surgical intervention is injection medialization. This can be performed as an office-based procedure using only local anesthesia, or in the operating room under general anesthesia. In either case, a variety of injectable materials are placed within the vocal fold lateral to the vocal process, to medialize the cord. In the past, Teflon was commonly used for this purpose, but has now largely fallen out of favor secondary to a high rate of complications. An alternative, long lasting (but not permanent) injectable is calcium hydroxyapetite microspheres in methylcarboxycellulose carrier gel (Radiesse Voice). Temporary injectables include gelfoam paste, hyaluronic acid, micronized cadaveric dermis (Cymetra), cross-linked collagen (Zyderm), and methylcarboxycellulose gel (Radiesse Voice Gel).

Surgical reinnervation for unilateral vocal cord paralysis is gaining popularity. Approaches can include nerve-muscle pedicle (using the omohyoid and ansa hypoglossi nerve) and direct nerve-nerve reinnervation (ansa cervicalis to recurrent laryngeal nerve). Results from these techniques are generally good, but can take 6 months or longer to be realized. For this reason, many surgeons combine reinnervation with injection medialization using a temporary substance.

For bilateral vocal cord paralysis, treatment approaches have a different perspective. Whereas restoration of voice is the primary goal for unilateral paralysis, resolution of potential or actual airway compromise is of tantamount importance in cases of bilateral paralysis. Most patients are initially treated with a tracheotomy to bypass the glottic obstruction. One common surgical procedure is lateralization of the vocal folds by arytenoidectomy. While this usually provides a patent airway and allows decannulation (reversal of the tracheotomy) the patient’s vocal quality usually suffers significantly.

Another surgical approach, partial posterior cordectomy) can often preserve vocal quality to some degree, while providing improved airway. This approach uses a laser through a surgical laryngoscope to remove a c-shaped wedge from the posterior portion of one vocal cord. This preserves a bilateral vibratory margin anteriorly while providing an airway posteriorly. The technique is often best performed as multiple less aggressive procedures to fine-tune vocal quality versus airway, rather than a single definitive procedure.

D. Recurrent Respiratory Papillomatosis

Recurrent respiratory papillomatosis is caused by the human papilloma virus, specifically types 6 and 11. Human papilloma virus is a small nonenveloped virus that infects the nuclei of host cells for replication. Papillomaviruses show a preference for infection of epithelial tissues and are very common in humans. Human papillomavirus (HPV) is responsible for a variety of disease including skin warts, recurrent respiratory papillomatosis, and invasive cancers such as cervical or oropharyngeal carcinoma. Transmission is hypothesized to involve vertical transmission during childbirth; maternal presence of condylomata during the perinatal period confers a 200-fold increase in relative risk for developing respiratory papillomatosis.

Benign, recurrent hyperplastic tissue growth of the upper airway characterizes this disease. It has a bimodal age distribution, with incidence peaks in children (infants to 12 years old), and in adults (30-40 years of age). Patients typically present with dysphonia or aphonia, although advanced disease can cause stridor from impending airway obstruction.

The primary site affected is the larynx, with the glottis being the most common area followed by the supraglottic larynx (Figure 15–10). Respiratory papillomatosis generally remains confined to the larynx but can spread distally to affect the trachea, bronchi, and lungs. Disease recurrence is frequent, and reports of children requiring more than 100 surgical procedures are not uncommon.

The primary therapy for respiratory papillomatosis remains surgical debulking, primarily by microdebrider or carbon dioxide laser ablation. Adjuvant medical treatments include cidofovir, indole-3-carbinol, ribavirin, mumps vaccine, and photodynamic therapy. The role for these adjuvant therapies in the treatment of respiratory papillomatosis is still being elucidated. Many otolaryngologists advocate avoidance of tracheotomy if possible, because this has been epidemiologically associated with increased risk of spread of papillomas to the lower respiratory tract. It is possible because patients with inherently more aggressive disease are also those who are more likely to require tracheostomy.

Pathophysiology of Head and Neck Squamous Cell Carcinoma

Excluding skin cancer, the vast majority of head and neck cancers are squamous cell cancers. This is the sixth most common malignancy; over 500,000 patients worldwide are diagnosed with head and neck squamous cell carcinoma each year. Head and neck squamous cell carcinoma is a potentially curable malignancy when diagnosed at an early stage. Unfortunately, patients often present with locally advanced disease. Overall, the prognosis is poor in this latter group of patients. For patients with advanced disease, between 60% and 70% develop loco-regional recurrences within 2 years.

Despite numerous advances in our understanding of this disease and in development of improved medical and surgical therapies, mortality rates have not changed significantly over the last two decades. Additionally, current therapies (both surgical and nonsurgical) carry a high level of treatment-related morbidity; patients with advanced-stage disease often suffer significant speech and swallowing impairment following treatment.

Development of head and neck squamous cell carcinoma is traditionally attributed to the synergistic carcinogenic effects of tobacco and alcohol. Other factors, such as poor oral dentition, gastroesophageal reflux, betel nut chewing, and viral infection may also be important contributors to the development of head and neck squamous cell carcinoma. It is now accepted that high-risk types (eg, 16, 18, and 31) of HPV are a causal agent in the development of some head and neck squamous cell carcinomas, particularly cancers of the oropharynx. While as many as 50% of oropharyngeal cancers have detectable HPV-16 DNA, only 25%-30% of all oropharyngeal cancers are thought to have a viral etiology (as opposed to the usual insults of tobacco and alcohol). In HPV-mediated transformation, oncoproteins (primarily E6 and E7) encoded by the HPV genome are transcribed by infected human epithelial cells, resulting in the degradation and inactivation of host tumor suppressor genes p53 and Rb. This is thought to allow unchecked cell cycle progression and development of genetic instability. Head and neck squamous cell carcinomas with active (eg, transcribed) HPV share a common molecular phenotype, similar to that of HPV-caused cervical cancer. Head and neck squamous cell carcinomas with inactive HPV appear to share a molecular phenotype in common with HPV-negative head and neck squamous cell carcinomas. The role, if any, of HPV in these latter cancers is under investigation.

Head and neck cancers are thought to progress through characteristic stages of premalignancy (dysplasia, carcinoma in situ, until finally reaching invasive carcinoma). This progression reflects the accumulation of multiple genetic changes from mutation, gene silencing (via methylation), chromosomal rearrangements, deletions and duplications. Common changes include mutational inactivation of p53 and Rb tumor suppressors, abrogation of the tumor suppressor p16 axis, amplification of the CCND1 locus (the gene encoding the cell cycle mediator cyclin D1), and various changes resulting in overexpression of epidermal growth factor receptor (EGFR), c-Met and transforming growth factor alpha (TGF-α).

Another important concept in head and neck cancer biology is that of field cancerization. According to this theory, first described in 1953, the entire mucosa of the upper aerodigestive tract is exposed to the same carcinogenic insults. Thus, premalignant changes are likely in a wide field, not just at the site of the original cancer. Even if that cancer is successfully treated, the patient remains at high risk for developing subsequent malignancies throughout the upper aerodigestive tract. Recent molecular-based investigations have provided strong support for this concept, and multiple studies have demonstrated genetic alterations in histologically normal tissue from high-risk individuals. For this reason, patients with a history of head and neck squamous cell carcinoma should be closely followed with yearly screening examinations for life.

Evaluation of the Patient With Head and Neck Cancer

A. History and Physical Examination

Any patient with a significant history of alcohol and tobacco use who presents with complaints related to the upper aerodigestive system should be evaluated with an index of suspicion for head and neck squamous cell carcinoma. Common symptoms of head and neck cancer can be quite subtle. Mechanical obstruction and dysfunction from tumor mass can produce dysphagia. Head and neck cancers are often associated with significant pain, and odynophagia is common for cancers of the oral cavity and oropharynx. Referred pain from neck metastases can result in otalgia, as can cancers arising in the nasopharynx or those affecting the EAC. Cancers of the true vocal cords often produce significant hoarseness early in the course of disease progression. Other symptoms can include hemoptysis, globus sensation, trismus, and weight loss. Sometimes the presenting symptom is a neck mass representing nodal metastases from an asymptomatic primary source.

Patients should be asked about family history of cancers and personal risk factors such as alcohol use and all forms of tobacco–cigarettes, cigars, chewing tobacco, etc. Certain ethnic populations (specifically emigrants from the country of India) may give a history of betel nut chewing. This nut has been found, like tobacco, to have synergistic carcinogenic risk when present in combination with alcohol consumption.

Physical examination should include a comprehensive evaluation of all sites of the head and neck. Inspection of the face and scalp for skin lesions, masses, and asymmetry should be performed. Otoscopy and anterior rhinoscopy allow evaluation of the EAC, tympanic membrane, and nasal cavity. Unilateral serous fluid in the middle ear space in an adult should raise suspicion for a nasopharyngeal lesion. The entire oral cavity should be evaluated, including each gingivo-buccal sulcus. The floor of mouth and base of tongue should be manually palpated for induration, pain on palpation, and asymmetry. Evaluation of the larynx and base of tongue can be performed by headlamp and mirror examination (indirect laryngoscopy). Flexible fiberoptic evaluation often provides additional detail and visualization not possible with the former method. Additionally, the nasal cavity and nasopharynx can easily be evaluated during the same procedure. A complete evaluation of the cranial nerves should be performed. This can often give insight into nerves affected by tumor mass effect (compression) or direct nerve invasion (perineural spread).

The neck should be carefully palpated for cervical lymphadenopathy. This is most easily accomplished by standing behind the patient and palpating each area of the neck symmetrically using both hands. This allows tactile comparison between sides. The pads of the fingers and thumbs are the most sensitive parts of the hands and should be used instead of the tips of the digits.

B. Imaging

Imaging is an important part of the diagnostic evaluation of a patient with head and neck cancer. Goals of imaging include preoperative identification of site and extent of tumor infiltration, presence of suspicious lymph nodes, and presence of anatomic variations. The imaging modality of choice for most suspected head and neck tumors is CT scan with intravenous contrast. This allows adequate soft tissue and bony detail. Most head and neck cancers will demonstrate enhancement on IV contrast administration, and the contrast additionally makes delineation of soft tissue structures much easier. In some cases contrasted MRI may offer additional information. These cases include evaluation of the skull base, parapharyngeal space and orbit, and evaluation of cranial nerves for signs of perineural spread.

Increasingly, evaluation of head and neck squamous cell carcinoma patients for distant metastases and monitoring for posttreatment disease recurrence or persistence involves the use of tumor imaging by 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) (Figure 15–11). The use of FDG-PET and combined PET/CT modalities is extremely important as early identification of persistent/recurrent disease has the potential to impact survival, especially for patients with advanced nodal disease. While CT and MRI rely on contrast enhancement patterns and differences in tissue attenuation, FDG-PET works somewhat independently of these criteria. FDG follows a cellular uptake pathway similar to glucose and becomes concentrated in cells with elevated glucose utilization. Because changes in tumor metabolism often precede changes discernable on clinical examination this makes FDG-PET potentially valuable in discovering clinically nondetectable tumor spread or recurrence. PET-CT results are generally reported as standardized uptake values (SUV). This is a ratio of tissue FDG uptake normalized to injected dose of FDG, patient body weight, and serum glucose levels. Higher SUV values indicate increased FDG uptake. Multiple tissue metabolic changes can lead to increased FDG uptake, including inflammation, infection, radiation response, and malignancy. While still a matter of debate, general consensus is that SUV greater than 2.5-3.5 should be considered suspicious for malignancy.

C. Staging and Treatment of Head and Neck Squamous Cell Carcinoma: TNM Staging

Staging for head and neck squamous cell carcinoma in the United States uses a TNM system. Staging is important because it allows estimation of prognosis, treatment planning, and expected response to treatment. The TNM system classifies cancers by T–primary tumor size, N–locoregional nodal metastases, and M–distant metastatic spread. The rules for classification vary by subsite within the head and neck. In general cancers of subsites that can be seen directly (ie, oral cavity, oropharynx, etc) are T-staged by size. By comparison, cancers of subsites not visible directly (ie, larynx, hypopharynx, sinonasal, etc) are T-staged by anatomic spread of disease.

Initial evaluation of the primary site and T-staging is best accomplished by operative direct laryngoscopy and esophagoscopy. This allows examination of the primary site and evaluation of the entire upper aerodigestive tract for simultaneous second primary tumors and biopsy of suspicious areas. Some authors propose trans-nasal esophagoscopy as an acceptable alternative to operative endoscopy. This method allows evaluation (and biopsy) of the most common areas for malignancy with the exception of the postcricoid region and subglottis. Transnasal esophagoscopy has the advantage of being able to be performed as an office-based procedure without the need for sedation.

It should be mentioned that cancers of Waldeyer’s ring which includes the palatine tonsil and base of tongue (lingual tonsil), while usually squamous cell cancer, can also be lymphoid in origin (such as lymphoma). If there is a suspicion for lymphoma, biopsy specimens should be sent as fresh material and not placed into formaldehyde. Often, flow cytometry and other pathologic tests useful for evaluation of lymphomas cannot be performed on formaldehyde-fixed specimens.

N-stage is determined by the presence of clinically evident regional lymph node metastases. It is important to point out that for head and neck cancer suspicious lymphadenopathy identified only on CT or MRI scan is included in the clinical staging. Radiologic findings suspicious for malignant lymph node metastases include lymph nodes that are round, heterogeneously enhancing, and larger than 1 cm. The current criteria for N-staging (for subsites other than nasopharynx) is summarized in Table 15–7.

M-stage is determined by the presence of distant metastatic disease. The most common sites of metastases for head and neck squamous cell carcinoma are the liver and the lungs. Therefore part of the initial staging workup for a patient with head and neck cancer includes some method of evaluating these structures. This most commonly consists of liver function blood tests and a chest roentgenogram, although at some institutions whole-body CT-PET scan is becoming more prevalent. M-stage is reported as Mx (cannot be determined), M0 (no distant metastases) or M1 (distant metastatic disease).

Overall staging is achieved by combining the T, N, and M stages to yield four broad stages (stage I-IV, with I carrying the best prognosis and IV the worst). Stage IV is further split into IV-a (simplified, denoting a surgically treatable cancer) and IV-b and -c (surgically unresectable). The overall staging system (based on the AJCC Cancer Staging Manual, sixth edition) for sites other than nasopharynx is summarized in Table 15–8 and Figure 15–12.

Therapy

Traditional therapy for head and neck squamous cell carcinoma depends on the stage of the cancer and varies by subsite. Simplified, early stage cancers are usually treated with single modality therapy, either surgical excision or radiation therapy. Advanced stage (stages III and IV) cancers are best treated by combined modality therapy, either surgery and postoperative radiotherapy, or chemotherapy/radiation therapy. Current research has shown that the latter is best delivered as concurrent chemotherapy-radiation therapy.

One new modality being investigated for head and neck squamous cell carcinoma includes molecularly targeted therapy. These therapies are designed to be specific for cancer cells, and include monoclonal antibodies and small molecule protein inhibitors. The most studied targeted therapy in head and neck squamous cell carcinoma is cetuximab, which targets the EGFR. In the initial randomized controlled clinical trial, researchers found that concurrent cetuximab and radiotherapy improved survival and locoregional disease control compared to radiotherapy alone. Studies have also demonstrated improvement in survival when cetuximab is combined with traditional chemotherapy. The EXTREME (Erbitux in First Line Treatment of Recurrent or Metastatic Head and Neck Cancer) trial randomized 442 patients from 17 countries to either chemotherapy (5-fluorouracil plus either cisplatin or carboplatin) or the same chemotherapy plus cetuximab. Patients in the cetuximab arm had 20% reduction in risk of death, and overall survival was increased from a median 7.4 months to 10.1 months. Other targeted therapies currently being studied include bevacizumab (an inhibitor of VEGF receptor, involved in angiogenesis) and various orally administered EGFR small-molecule tyrosine kinase inhibitors. Particularly promising are combination targeted therapies which include agents addressing both the EGF and VEGF pathways.

Cancers of the Oral Cavity

Patients with oral cavity cancer typically present with a painful mass of the tongue, buccal mucosa, floor of mouth, or alveolar ridge. Appearance to visual examination alone can be misleading. While most cancers have either an ulcerative or an exophytic appearance, many can have only subtle visually detectable changes. On palpation, the involved area is usually firm and indurated. A bimanual examination of the floor of mouth is mandatory for a complete examination. The mass should be manipulated to discern mobility. Oral cavity cancers can invade the mandible quite readily, and immobility should raise concern for bone invasion.

As with other head and neck subsites where the primary tumor can be directly visualized, T-staging is assessed by tumor size (summarized in Table 15–9). Cancers which invade adjacent structures are, in general, up-staged to T4 regardless of size. Note that superficial erosion of the bony tooth socket by a gingival primary tumor is not sufficient to upstage to T4.

The nodal drainage patterns of oral cavity cancers are of particular concern, because the incidence of occult metastases can approach 30% depending on T-stage and depth of invasion. Cancers of the upper lip can drain to the parotid bed in addition to level I. When planning treatment of occult metastases, it is important to remember that midline cancers of the oral cavity often drain bilaterally.

Treatment for oral cavity cancer is primarily surgical. Cancers of the lip can often be addressed with simple wedge excision, with 5-mm borders of surrounding normal tissue. If less than one-third of the lip is excised, a reasonable result can be obtained with simple closure. Larger excisions require a reconstructive procedure, often involving transfer of lip tissue from the contralateral lip. Cancers of the tongue, floor of mouth, and buccal mucosa are excised with a margin of 1 cm normal appearing tissue. Often primary closure, with or without a split thickness skin graft, will provide acceptable functional outcomes. Large resections usually require a flap reconstruction, especially if bone is involved, and a segmental mandibulectomy needs to be performed. The flap reconstruction can be either a pedicled flap (such as pectoralis major flap), or a free flap (radial forearm and fibular being common flap harvest sites).

Assuming appropriate treatment, outcomes for patients with oral cavity cancer are generally good. The prognosis for squamous cell carcinoma of the lip is excellent, with 5-year survival of 91% for stage I-II, 83% for stage III-IVb, and 52% for stage IVc. Squamous cell carcinoma of the oral cavity proper carries a reduced, yet still relatively good prognosis. Five-year survival rates are 72% for stage I-II, 44% for stage III-IVb, and 35% for stage IVc.

Cancers of the Oropharynx

The oropharynx consists of the soft palate, base of tongue, palatine tonsils, and posterior/lateral oropharyngeal walls. Patients with cancer of the soft palate usually present with relatively early stage lesions, which typically remain fairly superficial. They usually arise on the anterior aspect of the palate and patients may report noticing a lump in the palate while swallowing. This contrasts with cancers of the tongue base, tonsillar region and pharyngeal walls, which typically present at advanced stage. These anatomic regions lack the rich supply of pain nerve fibers typical of the oral cavity and soft palate, and cancers can spread significantly before being noticed. Typical presenting symptoms usually are attributable to sequelae from invasion of adjacent structures, and can include dysphagia, cranial nerve defects, referred otalgia, trismus (from pterygoid muscle invasion), or neck mass (from nodal metastases). Because early-stage oropharyngeal cancers often are asymptomatic, any erythroplastic or suspicious lesion in these areas should be biopsied even if no associated symptoms are present.

As is the case for oral cavity cancer, T-staging for squamous cell cancer of the oropharynx is generally determined by tumor size (summarized in Table 15–10). Again, invasion of adjacent structures results in classification as T4 regardless of size. Oropharyngeal cancer N-staging is the same as other head and neck squamous cell carcinoma subsites (except nasopharynx). Nodal metastases from oropharyngeal cancers are common, with 70% of patients with oropharyngeal cancer having ipsilateral cervical nodal metastases at the time of presentation. Bilateral cervical nodal metastases are also relatively common, up to 30%-50%, depending on size and subsite of the primary tumor.

Treatment for oropharyngeal cancer, especially the tongue base and soft palate, is often weighted toward non-surgical therapy. This is because surgical resection can result in extensive morbidity in terms of velopharyngeal insufficiency (for palate) and dysphagia (for tongue base). Because of interruption of the blood supply to the remainder of the tongue, large base of tongue cancers usually require total glossectomy even if the anterior tongue is spared. Surgical resections of all but the smallest oropharyngeal cancers usually require flap reconstruction; in the modern era this is most often in the form of a free flap with microvascular anastomoses. A tracheotomy is usually performed during the initial surgical resection for maintenance of adequate airway. As collective experience with robot-assisted oropharyngeal surgery advances, surgical treatment is likely to become more common as this approach appears to have decreased morbidity compared to open techniques.

As with other head and neck squamous cell carcinomas, early stage (I-II) cancers are treated with single modality therapy, and advanced stage (III-IV) cancers require multiple modalities. It should be noted that nonsurgical modalities have their own set of consequences, and there is some evidence demonstrating worse swallowing outcomes following chemotherapy-radiation compared to surgery and postoperative radiotherapy for advanced stage oropharyngeal lesions.

The association between HPV and oropharyngeal cancers deserves special mention. HPV is the accepted etiologic cause of greater than 95% of cervical cancer. Since the first demonstration of HPV in a head and neck squamous cell carcinoma tumor in 1985, numerous studies have found HPV DNA in head and neck squamous cell carcinomas from various subsites, most commonly the oropharynx. It now appears that of oropharyngeal squamous cell carcinomas, approximately 50% have no association with HPV, 25% are likely to be caused by HPV (similar to cervical cancer), and 25% have HPV DNA but an unclear relationship between the virus and the cancer. Future studies assessing the role of HPV in this latter group are currently being developed.

Survival for oropharyngeal cancer is generally worse than that of oral cavity cancer. Five-year survival for stage I-II oropharyngeal cancer is 58%, stage III-IVb is 41%, and stage IVc only 20%.

Cancers of the Larynx

Interestingly, laryngeal cancer was an extremely rare disease until the 20^th century. Mass-produced cigarettes came into vogue in the 1900s; soon after this a dramatic rise in the incidence of laryngeal cancer began to be noted. Tobacco exposure is now accepted as the primary etiologic agent responsible for laryngeal cancer. Other possible additive factors include laryngo-pharyngeal reflux and possibly certain viruses such as herpes simplex or HPV.

Laryngeal cancers can arise from above the true vocal folds (supraglottic region), below the true vocal folds (subglottic region) or from the true vocal folds (glottic region). The latter represents greater than 75% of all laryngeal cancers. For patients with glottic laryngeal cancer, dysphonia (hoarseness) is the most common presenting symptom. Any hoarseness that persists longer than 2 weeks, especially in a patient with associated risk factors of tobacco or alcohol use, should undergo laryngeal examination. Patients with hoarseness, however, are more likely to have a non-malignant cause for their symptoms than to have laryngeal cancer.

Laryngeal cancers arising from the subglottis or supraglottis usually present at a later stage than do glottic cancers, as hoarseness does not develop until late in the disease process. Oftentimes, the presenting symptom can be life-threatening stridor from an obstructive mass. Emergent awake tracheotomy is sometimes required to secure a stable patent airway in these cases. Other symptoms can include globus sensation (“lump in throat”) and dysphagia. Evaluation for preexisting swallowing dysfunction is emerging as an essential part of the diagnostic workup for laryngeal cancers. Because of the emphasis on organ-preservation strategies, identification of patients likely to end up with a nonfunctioning larynx may alter treatment planning. Currently this is an area of active investigation.

Unlike oral cavity and oropharyngeal cancer (T-staged by tumor size), cancers of the larynx are T-staged by anatomic spread of disease (Table 15–11). One important factor is determination of vocal cord fixation. This can be reliably determined with greater than 90% accuracy based on in-office flexible fiberoptic evaluation combined with high-resolution CT imaging of the larynx. Laryngeal cancer N-staging follows the convention for other head and neck subsites (except nasopharynx).

Therapy for early (T1-T2) glottic cancer has traditionally been radiation therapy, especially for early, diffuse disease involving both true vocal folds and the anterior commissure. In these cases, surgical resection would result in significant disruption of the normal vocal fold architecture and function. This should be balanced with the knowledge that radiation therapy is essentially a one-time treatment modality, which carries its own spectrum of side effects and sequelae.

The goal of organ preservation (ie, preserving a functional ability to phonate and protect the airway during deglutition) should be kept in mind when planning treatment options. In the now famous “fireman’s study,” McNeil et al found that many people would accept a 20% decrease in survival, rather than lose their larynx. Surgical organ preservation strategies include endoscopic microsurgical excision, supracricoid laryngectomy, and vertical partial laryngectomy. Endoscopic microsurgical excision can allow maximal preservation of vocal function in selected T1 glottic cancers, while maintaining a sound oncologic outcome. In this approach, dissection is meticulously performed just deep to the most involved layer of the true vocal fold (epithelium, superficial lamina propria, deep lamina propria or vocal ligament, vocalis muscle). Partial laryngectomies (vertical partial and supracricoid) can yield satisfactory voice and swallowing function, but require extensive patient cooperation in the postoperative period. Supracricoid laryngectomy involves removal of the supraglottis, false and true vocal cords, and thyroid cartilage. The hyoid bone, cricoid cartilage, and at least one arytenoid are preserved. The movement of the remaining arytenoid against the tongue base allows phonation and (eventual) preservation of swallowing function. Postoperatively, aspiration is to be expected and patients with poor preoperative pulmonary function should not be offered this option. Total laryngectomy is the treatment of choice for salvage of organ-preservation failures, as well as the primary treatment for most advanced glottic (T3 and T4) cancers.

Outcomes for patients with laryngeal cancer can be excellent. For early stage I glottic carcinoma (tumor limited to the true vocal folds), 5-year overall survival rates of 90% can be expected. Overall, laryngeal cancer carries a 5-year survival for stage I-II disease of 79%. With invasion of adjacent structures characteristic of stage III-IVb disease, 5-year survival decreases to 55%, and with distant metastases (stage IVc) the 5-year survival is 35%.

Cancers of the Hypopharynx

The hypopharynx is a continuation of the oropharynx, and extends superiorly from the level of the hyoid bone to the level of the inferior aspect of the cricoid cartilage. It consists of the pyriform sinuses (the most common subsite for hypopharyngeal cancer), the posterior pharyngeal wall and postcricoid regions. The hypopharynx is lined with stratified squamous epithelium and is invested with an abundant lymphatic drainage network. Patients with hypopharyngeal cancer typically present with advanced stage disease, stage III or worse. There are several contributing factors, including lack of specific symptoms for small lesions in this region, and a lack of anatomic boundaries to prevent spread of disease. Typical presenting symptoms can include referred otalgia, odynophagia, and dysphagia. Other symptoms can include a neck mass (from nodal metastases), hoarseness (due to laryngeal involvement), and weight loss. Like laryngeal cancer, hypopharyngeal cancer is chiefly related to excessive alcohol and tobacco exposure. There is also some evidence for a possible contribution of GERD to hypopharyngeal cancer. This may explain the increased likelihood for patients with Plummer–Vinson syndrome to develop hypopharyngeal cancer, regardless of tobacco or alcohol exposure. In Plummer–Vinson syndrome, patients develop esophageal webs, and chronic acid exposure above the site of webbing is thought to possibly result in chronic inflammation predisposing to malignancy.

Hypopharyngeal cancer T-staging is somewhat unique, as it is a combination of tumor size (as for oral cavity and oropharyngeal tumors) and anatomic spread (as for laryngeal cancer). The T-staging criteria for hypopharyngeal cancer are summarized in Table 15–12. N-staging and M-staging are the same as for the majority of head and neck sites (except nasopharnyx).

Early stage (T1-T2 tumors) hypopharyngeal cancer is often treated with primary radiotherapy. Surgical resection is more likely in these instances to lead to dysphagia and aspiration. Alternatively, laryngeal conservation surgery may be a reasonable option for some of these early-stage cancers. More advanced (T3-T4) tumors can be treated by surgical resection with postoperative radiotherapy, or chemotherapy/radiation therapy. The standard surgical procedure is total laryngectomy with partial pharyngectomy. Tumor extension into the esophagus requires a cervical esophagectomy. The resulting alimentary tract defects require repair by various methods, including microvascular jejunal free flap, gastric pull up, or microvascular myocutaneous flap (most often tubed radial forearm or anterolateral thigh). Organ preservation surgical strategies can achieve laryngeal preservation rates of up to 40% for highly selected cases, but these remain somewhat controversial. As previously described for laryngeal cancer, preoperative evaluation of a patient’s pulmonary function status is critical, as significant postoperative aspiration can be expected. Because of the high incidence of occult nodal metastases, treatment of the neck lymphatics should be included in any therapeutic plan. For clinically evident neck disease, neck dissection is indicated. For patients with a clinically negative neck, optimal management is still being elucidated. Elective neck dissection represents one option for the clinically negative neck. Alternatively, if radiation therapy is planned for treatment of the primary site, the radiation fields can be adjusted to include treatment doses to the neck lymphatics.

Hypopharyngeal cancer carries the worst prognosis of any head and neck subsite. The underlying reasons for this are still unclear, and several hypotheses have been suggested. A distinctive pathologic feature of hypopharyngeal cancer is the propensity for submucosal spread of disease, often resulting in clinically understaging disease extent. The abundant lymphatic drainage of the hypopharynx may also predispose the patient to early nodal metastases. It has also been suggested that the lack of definitive anatomic boundaries may allow early spread of disease. Given all of these factors, it should not be surprising that 5-year survival for stage I-II hypopharyngeal cancer is 47%, stage III-IVb is 30%, and stage IVc only 16%.

Salivary Gland Neoplasms

Neoplasms (both benign and malignant) of the salivary glands are relatively uncommon, representing approximately 2% of all head and neck neoplasms. In general, the larger the salivary gland, the higher the overall incidence of neoplasms—but these neoplasms are also less likely to be malignant. Major salivary glands include the parotid, submandibular, and sublingual glands. The minor salivary glands are scattered mainly throughout the mucosa of the lips and oral cavity, and to a lesser extent the entire upper aerodigestive tract. For these smaller glands, the incidence of neoplasms is lower, but if one does occur it is more likely to be malignant. The parotid gland accounts for 80% of all neoplasms, while 15% arise from the submandibular glands and 5% from the sublingual and minor salivary glands. For any given tumor, the incidence of malignancy is 20% for parotid tumors, 60% for submandibular tumors, and 80% for sublingual and minor salivary gland tumors. Unlike squamous cell carcinoma, tobacco and alcohol exposure do not appear to play an etiologic role in salivary gland malignancy. It should be noted that the parotid gland contains lymph nodes that drain the preauricular and temporal skin; any parotid mass should raise suspicion for a skin cancer with intraparotid nodal metastases and a complete examination of the scalp and preauricular skin is indicated.

The most common benign salivary gland tumor is pleomorphic adenoma (also called benign mixed tumor), representing 60% of all salivary gland neoplasms. Patients typically present with a slowly enlarging, painless firm mass (most often in the parotid gland). Facial nerve paralysis is exceedingly rare, even with very large tumors, due to the slow-growing nature of these neoplasms. Malignant transformation of pleomorphic adenoma is rare but does occur; the resultant malignancy is highly aggressive. Treatment for pleomorphic adenoma is complete surgical excision including a surrounding margin of normal tissue. Due to microscopic transcapsular tumor infiltration, simple enucleation is insufficient for these benign neoplasms.

The second most common benign salivary neoplasm is papillary cystadenoma lymphomatosum (also known as Warthin tumor). This benign tumor is responsible for 10% of parotid neoplasms, but is rare outside of the parotid gland. Bilateral tumors are present in up to 10% of patients so careful palpation and/or imaging of both parotid beds is important if this tumor is suspected. Patients typically present with a slowly enlarging parotid mass, which is painless and rubbery upon palpation. Standard treatment is complete surgical excision (such as superficial parotidectomy with facial nerve preservation) although some authors advocate that enucleation of these tumors may be adequate therapy.

The most common salivary gland malignancy is mucoepidermoid carcinoma. Despite the propensity of the smaller salivary gland neoplasms to be malignant, the most common site for mucoepidermoid carcinoma is still the parotid gland. Approximately 45%-70% of mucoepidermoid carcinomas arise from the parotid gland, while 20% occur in minor salivary glands of the palate. Typical symptoms at presentation can be quite similar to those of benign salivary neoplasms, with a painless enlarging mass. Symptoms such as pain or facial paralysis are uncommon, but when present should raise suspicion for a high-grade aggressive lesion. Mucoepidermoid cancers are graded as low grade (mostly mucus cells), intermediate grade, and high grade (characterized by a hypercellular solid tumor). The latter can be difficult to distinguish from squamous cell carcinoma without immunohistochemical staining. Treatment is tailored to both the extent of disease, tumor grade, and location. Localized disease is usually amenable to surgical excision (parotidectomy with facial nerve preservation, submandibular gland excision, or wide local excision for minor salivary gland origin). Advanced disease requires extensive resection, often combined with neck dissection and/or postoperative radiotherapy.

The second most common salivary gland malignancy is adenoid cystic carcinoma. It is most common in the submandibular, sublingual, and minor salivary glands. Of all parotid neoplasms, adenoid cystic carcinoma is more likely to present with pain or paresthesias, although an asymptomatic mass is still the more common presentation. Adenoid cystic carcinoma is characterized by a propensity for perineural spread. Thus, preoperative imaging with gadolinium-enhanced MRI is often helpful in therapeutic planning. Treatment consists of complete surgical resection (sometimes requiring facial nerve sacrifice) and postoperative radiation therapy. There is some evidence that postoperative proton-beam therapy offers improved outcomes compared to traditional radiation therapy, although the former is available at only a few centers in the United States. Adenoid cystic carcinoma rarely spreads to cervical lymphatics, so neck dissection is not routinely advocated. It does, however, have a propensity for distant metastases—especially the lung. The disease is also characterized by frequent local recurrence, up to 40% even after adequate local excision. Because of the relatively slow growing nature of this disease, survival for adenoid cystic carcinoma at 5 years is favorable, approximately 65% for all stages. Due to the propensities for local recurrence and distant metastases, however, this statistic decreases to 12%-15% at 15 years.

As mentioned above, malignant degeneration of pleomorphic adenomas can occur although it is uncommon. Some authors have estimated the risk of malignant degeneration at 1.5% for the first 5 years, increasing to 9.5% by 15 years or longer. The resulting malignant tumor is termed “carcinoma ex-pleomorphic adenoma” and is characterized by an aggressive natural history and poor clinical outcome. The most common presentation is sudden rapid growth of a previously stable parotid mass. Histologically, the malignant cells can take the form of any epithelial malignancy except acinic cell. Carcinoma ex-pleomorphic adenoma is characterized by frequent lymph node metastases, and up to 25% of patients will have clinically evident cervical lymphadenopathy on presentation. Treatment includes radical surgical resection, often combined with neck dissection and/or postoperative radiotherapy.

The prognosis of salivary gland malignancies varies according to histologic type, location, extent of disease, and grade. The criteria for T-staging are similar to other head and neck sites with directly measurable tumor size, and are summarized in Table 15–13. Overall, early stage (stage I-II) salivary gland cancers carry an excellent prognosis, greater than 80% at 5 years. Advanced (stage III-IV) salivary gland cancers on the other hand, have overall 5-year survival rates ranging from 23% to 56% depending on type and grade.

References