19 Neurological Voice Disorders

The word vagus comes from the Greek “wanderer,” and it is indeed the longest of the cranial nerves with the most wandering course. It provides both afferent and efferent fibers. About 80 to 90% of nerve fibers in the vagus nerve are afferent. Afferents include the somatic sensory fibers from the skin of the ear, the posterior ear canal, and the tympanic membrane. General afferent and efferent visceral fibers include those from the heart, pancreas, stomach, esophagus, upper respiratory tract, and pharynx. There are afferent autonomic fibers and afferent taste fibers.1,2

Efferent fibers include the somatic motor fibers to the vocal fold, larynx, and pharynx.

The vagal afferent central fibers originate in the nodose and jugular ganglions, which are sensory ganglions. They enter the medulla oblongata and lie in the groove between the olive and inferior peduncle with nerve roots of IX and XI. Rootlets unite to pass beneath the cerebellar flocculus. The vagus nerve leaves the cranial vault with cranial nerves IX and XI in the jugular foramen.³

Taste fibers from the epiglottis and larynx pass up the vagus to join the tractus solitarius. They terminate in the nucleus tractus solitarius. There are contacts with motor centers of medulla oblongata, pons, and spinal cord for mastication and deglutition.

Afferent fibers from the supraglottic larynx are transmitted by the internal branch of the superior laryngeal nerve. Those from the glottis and subglottis are transmitted by the recurrent laryngeal nerve. These go proximally to the nodose ganglion and then to the nucleus tractus solitarius, on an unilateral basis. There are then bilateral axonal impulses to the nucleus ambiguus at which time bilateral efferent fibers can, for example, lead to a bilateral glottic closure reflex arc.

A little more information on neural function to the larynx taken from animal research reveals the following: In squirrel monkey and rat, viscerotopic proprioceptive feedback has been demonstrated from the lungs and larynx to the nucleus tractus solitarius.

In squirrel monkey, the periaqueductal gray and lateral tegmentum of the midbrain have been noted to be crucial for vocalization. They appear to collect auditory, visual, and somatosensory input from diverse processing structures, motivation controlling input from limbic structures, and volitional input from anterior cingulate cortex. These are then connected with phonatory motor neurons via single to multiple interneurons. The nucleus retroambiguus has been noted to be a premotor relay station for laryngeal and expiratory vocalization.

Laryngeal receptors have been noted to be denser on the laryngeal surface of the epiglottis than on the true vocal folds and denser on the posterior than on the anterior aspects of the true vocal folds. Chemoreceptors appear to be limited to the supraglottis. They respond to changes in pH and water and appear able to detect chemical nonsaline-like molecules. Chemoreflexes include vocal fold closure, laryngospasm, apnea, bradycardia, hypertension, cough, and bronchoconstriction.

Peripherally, the velocity of axonal transmission of the recurrent laryngeal nerve is approximately 60 m/s. The recurrent laryngeal nerve supplies all intrinsic muscles except the cricothyroid muscle (supplied by the external branch of the superior laryngeal nerve). The posterior cricoarytenoid muscle is said to be the only abductor of the vocal folds. It is supplied by the recurrent laryngeal nerve. The only intrinsic muscle of the larynx said to have bilateral motor innervation is the interarytenoid muscle, supplied by both recurrent laryngeal nerves.

The course of the left recurrent laryngeal nerve is longer than that of the right, but the maximum diameter of fibers and of the myelin sheath of the left is slightly greater than that of the right, thereby allowing for approximately simultaneous activation of bilateral laryngeal muscles.

Extracranially, the cranial nerve XI joins the vagus near the nodose ganglion to supply motor branches. The vagus nerve trunk runs in the carotid sheath deep to and between the internal jugular vein and carotid artery.

The left vagus nerve enters the thorax between the aorta and the left pulmonary artery. It gives off the left recurrent laryngeal nerve that loops over the aortic arch distal to the ligamentum arteriosum and ascends in the tracheoesophageal groove medial to the thyroid gland, under the inferior constrictor muscle. It enters the larynx through the cricothyroid membrane behind the cricothyroid joint. The right vagus nerve crosses superficial to the subclavian artery. It gives off the right recurrent laryngeal nerve that loops under the subclavian artery and ascends in the tracheoesophageal groove in a more angulated course than the left recurrent laryngeal nerve. Approximately 1% of individuals have a nondescending right recurrent laryngeal nerve, due to the anatomical variant of a retroesophageal subclavian artery. Such individuals may also have dysphagia “lusoria” due to this aberrant positioning of the artery.

The superior laryngeal nerve arises below the nodose ganglion. Its internal branch is sensory. It pierces the thyrohyoid membrane in close association with the superior laryngeal artery. It supplies parasympathetic secretory innervation to glands supplying the true vocal folds as well as general sensory innervation to the supraglottic larynx. The external branch supplies motor innervation to the cricothyroid muscle and the inferior constrictor muscle. Terminal branches of the superior laryngeal nerve communicate with the recurrent laryngeal nerve as the ansa galeni.1–7

Language and Central Connections for Voicing

The left hemisphere is dominant in 95% of the population; it is said to be more analytic than the right hemisphere and tends to analyze information collected by the right hemisphere. It deals more with language, understanding, and expression. The left temporal and frontal areas are particularly important for language. There are connections with deeper structures processing sensory data, especially touch and hearing. Analysis of word meaning occurs in Wernicke area: it demonstrates activity, for example, when consonants are heard. Broca area abuts on/instructs the motor cortex relating to articulation and speech.⁸

On positron emission tomography scans, word reading activates parts of the visual cortex. Listening to speech causes activity in the auditory cortex. Thinking about words causes Broca area to activate. Thinking about words and speaking generates widespread activity.

The right hemisphere has roughly equivalent areas for environmental noise, prosody, and spatial skills. It tends to deal with visual activities, is more “intuitive,” puts information together, and groups it.

As noted above, there are central connections of voicing involving many layers of the brain and brainstem. In the brainstem, the nucleus ambiguus is of particular importance from an effector standpoint and the solitary nucleus as an afferent center. In the midbrain, the periaqueductal gray and “limbic” system have a significant role. There are multiple connections in the subcortex. In the cortex areas such as Broca area, the motor and premotor cortex, the fronto and parietal cortex, and the anterior cingulate gyrus are all important. To look at it from a different perspective, a vocal idea might originate in Broca area, instructions might be sent to the motor nuclei in the cortex and precentral gyrus; coordinated muscular activity will then occur involving the larynx, thorax, abdomen, and articulators. Refinements occur via the extrapyramidal system, the basal ganglion, cerebellum, and the autonomic nervous system. Of course not only sound but also vibratory and tactile sensations are produced, which feed back among other systems through the auditory cortex. And this is vastly oversimplified.

Jürgens has identified in mammals a subcortical network of brain regions dedicated to phonation controlling many species-specific calls. He has postulated that the periaqueductal gray assists or regulates neurons of the lower brainstem via the reticular formation and lateral pontomedullary and limbic regions with cortical laryngeal projections. In humans, he postulates that this visceromotor pathway is exploited during emotive voicing, but that linguistic and paralinguistic speech involves more direct corticobulbar projections to the reticular formation. He emphasizes the temporal lobe activity for auditory self-regulation.^3–7,9

Neurological Voice Disorders: Overview

The “classic” classification in relation to neurological damage differentiates lower motor from upper motor neuron deficits, working from peripheral to central. Thus, injuries at the level of the muscle, neuromuscular junction, the peripheral nerve, and the brainstem nucleus all lead to a flaccid lower motor neuron deficit. Lesions more proximal to this tend to cause a spastic upper motor neuron deficit.¹⁰

In this type of classification, there are several other disorders that fall under categories such as dyskinetic/extrapyramidal (Parkinson disease, essential tremor, dystonia, myoclonus, etc.); ataxia/cerebellar (degenerative disorders, hemorrhage, ischemia, etc.); apraxia of the cortex or subcortex (trauma, stroke, tumor, cerebral palsy, etc.); and mixed (amyotrophic lateral sclerosis [ALS], multiple sclerosis, Shy-Drager syndrome, etc.).

In our voice clinic, we prefer to use the Ramig classification¹¹ which looks at neurological disorders as they effect the voice from the standpoint of hypoadduction/hyper-adduction/maladduction/abduction/mixed/miscellaneous. We find this helpful because it focuses on the effect of the disorder at a laryngeal level. (For swallowing issues, we do not find it as advantageous.)

Using the Ramig system, disorders of hypoadduction refer to any component of the motor unit (muscle, neuromuscular junction, peripheral nerve, or nucleus). Thus, they include all lower motor neuron disorders. They also include some cases of central origin including many cases of Parkinson disease, multiple system atrophy, traumatic and/or closed head brain injury, and trauma.

Hyperadduction refers to pseudobulbar issues, Huntington chorea, and adductor spasmodic dysphonia.

Maladduction refers to abductor spasmodic dysphonia and several paradoxical conditions.

Long-term tremor is seen in essential tremor, Parkinson disease, and cerebellar/dystonic tremor. Mixed disorders and miscellaneous include conditions such as multiple sclerosis (MS), ALS, apraxia, and Tourette syndrome.

Hypoadduction

A typical deficit at the muscular level might be a myopathy such as polymyositis or dermatomyositis. These are fairly unusual and are more likely to present with swallowing rather than pure voice issues. Muscular dystrophy and inclusion body myositis may fit into this category especially with oropharyngeal signs of progressive dysphagia and possibly with velopharyngeal insufficiency.12,13

The principal example of a disorder of neuromuscular junction transmission is myasthenia gravis. Myasthenia gravis^14,15 is a flaccid, fatigable condition, with a female to male prevalence of 2:1. Although its primary presentation is said to be ocular, approximately 30% will present initially with oropharyngeal symptoms. Thus, they may initially be seen in the voice clinic with symptoms such as vocal fatigue, dysarthria, dysphagia, and/or dysphonia progressing as the day progresses. To complicate matters, anticholinesterase receptor antibodies may be negative. Other investigations such as single-fiber electromyography (EMG) may also be negative and an empiric trial of pyridostigmine may be required. The Tensilon test, previously the main investigative test, is less used by many neurologists due to potential cardiac complications. Immunomodulatory therapy with corticosteroids, immunoglobulin, or plasma exchange in addition to cholinesterase inhibitors is the mainstay of treatment.

A lesion of the peripheral nerve could be caused by disorders such as trauma, tumor, iatrogenic illness, viral infection, collagen vascular disease, and Guillain-Barre syndrome. Terris et al¹⁶ published a review of recent literature in 1992 of over 1000 cases with unilateral vocal fold paralysis. In their series, 35.8% were caused by neoplasm (54.8% lung cancer); a further 24.6% were postsurgical, the majority being post-thyroidectomy. Recently, for several reasons perhaps including more thyroidectomy being performed in major centers and more being performed with nerve stimulators, the likelihood of a peripheral nerve injury involving the recurrent laryngeal nerve being caused by thyroid surgery is decreasing.

By and large, unilateral recurrent laryngeal nerve injury paralysis is easy to identify with an immobile vocal fold typically in the paramedian position. Unilateral superior laryngeal nerve injury has a more complex presentation. Causes are typically surgical trauma, blunt neck trauma, or viral. Clinical presentation is subtle with vocal fatigue and relatively mild voice complaints. We feel that it is likely that this type of injury is going underdiagnosed.

A lesion at the brainstem nucleus is typically caused by infarction, not uncommonly of the posterior inferior cerebellar artery. Typical signs in a “classic” presentation of a Wallenberg syndrome are an ipsilateral Horner syndrome, facial dysesthesia, and limb ataxia with contralateral body pain/temperature impairment and motor weakness. Symptoms from a laryngeal standpoint are dysphagia, dysphonia, and dysarthria. Other causes include Arnold-Chiari malformation, syringobulbia, tumor, or trauma.

It is not uncommon for Parkinson disease to cause a clinical picture of hypoadduction of the larynx.^11,17,18 Parkinson disease is a degenerative disorder of nigrostriatal extrapyramidal neurons which leads to a deficiency in dopamine. It is a rather common disorder affecting approximately 1:100 population over the age of 60 years and 1:1000 population under the age of 60 years. There are approximately 1,500,000 sufferers in the United States. Typical global symptoms include shuffling gait, muscle rigidity, resting “pill-rolling” tremor, flat facial affect, stooped posture, and impaired postural reflexes.

Management is symptomatic with dopaminergic agents. Unfortunately, this has no effect on the underlying neurodegenerative process, and symptom progression (along with potential complications of treatment) will be seen with the passage of time. Nonpharmacological treatments include neurosurgical deep brain stimulation for selected patients.^19,20

From the standpoint of voice, most individuals affected with Parkinson will develop some type of voice disorder (70 to 90%), and in early stages, voice and swallowing issues may be presenting signs. Symptoms include a soft, monotonous, often breathy voice with reduced volume and range. Vocal tremor may be present. Signs include bowed thin vocal folds with incomplete apposition on phonation. Articulation may also be impaired and sluggish, particularly involving lingual and labial consonants. On EMG, it is stated that there is reduced activation of the thyroarytenoid muscle, due to reduction of the central drive of the laryngeal motor neuron pool or due to loss of reciprocal suppression of the thyroarytenoid during inspiration. The Lee Silverman voice treatment has proven successful in several cases.

Parkinson plus syndromes generally include aspects of autonomic and/or supranuclear failure. These include syndromes such as progressive supranuclear palsy and multiple system atrophy. The voice is often breathy and can occur with unilateral or bilateral posterior cricoarytenoid dysfunction. Stridor and sleep disorder may be a relatively early symptom and must be watched for and managed aggressively—particularly in multiple system atrophy where it may contribute to nocturnal sudden death.^10,21–26

Hyperadduction of the vocal folds can be present in many neurological disorders. They can often be divided into constant and irregular.²⁷

Upper motor neuron deficits, sometimes called pseudobulbar palsy, typically present with fairly constant strained quality to the speaking voice.¹¹ Pseudobulbar palsy is an inaccurate term. The pathological findings are found bilaterally in the corticobulbar pathways of the pyramidal tracts at sites rostral to the cranial nuclei—especially the internal capsule. The most common cause is stroke, but it can be due to neurodegenerative (such as ALS), metabolic, and inflammatory conditions as well as neoplasms. The clinical picture is one of speech dysarthria (according to Duffy,^11,28 this is seen in 57% of stroke patients) and emotional lability. Other symptoms include dysphonia and dysphagia, while physical findings include a brisk jaw jerk and frontal release signs. The voice symptoms need to be reviewed in the context of the overlying speech dysarthria. The voice is often harsh, hoarse, monopitched and/or low pitched with a strained–strangled quality. There may also be pitch breaks. Depending on the severity of the underlying condition, the voice may also be hypernasal with air escape into the nose.

Adductor spasmodic dysphonia is another hyperadduction movement disorder.29–31 It is a focal, slowly progressive dystonia involving certain laryngeal muscles but reflecting central motor processing issues/abnormalities. Present only during certain specific speech vocal tasks, it can be overridden by vegetative phenomena such as laughing or by chanting or singing. It is not present during sleep. There is a background of normal speech overlain by vocal spasms that are not under voluntary control. This leads to a strained and strangled speech pattern. (See Chapter 21 on dystonias for further information on spasmodic dysphonia.)

Incidence estimates vary, with one suggesting an incidence of 1:100,000. The disorder is more common in women than in men. The most common form (in 85 to 90%) is adductor spasmodic dysphonia, although abductor and mixed forms do occur. It is characterized by pitch breaks associated with vowels, in particular words ending with a vowel and being followed on by a word starting with a vowel (for example, “we—eat”) and words with two vowels in tandem (for example, eight, year, eels). The vowels/e/and/a/are particularly problematic. Most common management is injection of low-dose botulinum toxin into the thyroarytenoid muscles.^32–38

Tremor is another form of hyperfunctional voice disorder. It is pervasive in neurological disease but also present in the normal neurological system (as physiological tremor). It can occur as part of spasmodic dysphonia. There are also familial forms of tremor, predominantly with autosomal dominant pattern of inheritance. Tremor is the most common form of movement disorder, with an incidence said to be between 4 and 60 per 1000. While hands and feet are most commonly affected, the larynx is affected in between 4 and 20% of cases.^39–41

Tremor can also be seen in conditions such as Parkinson disease (Logemann et al noted 13.5% of patients with vocal tremulousness), ataxic dysarthria, and palatopharyngeal myoclonus.¹⁷

Hyperadduction can occur with Huntington chorea. This is an autosomal dominant disorder affecting 4 to 8 per 100,000 population. There is a general loss of neurons in the caudate nucleus. Globally, the affected individual will have choreic movements affecting any part of the body. From a laryngeal perspective, irregular pitch fluctuations and voice arrests are seen. The voice tends to be harsh, with excess loudness, variations, and a strained–strangled quality.^27,42,43

When examination of the larynx demonstrates hyperfunction of the supraglottic structures and hyperadduction is being considered, it is most important for the examiner to consider the possibility of underlying hypofunction at the level of the true vocal folds, with secondary supraglottic hyperfunction.

A variety of other neurological disorders can manifest with vocal issues.^10,44–48 These include (among others) ataxic dysphonia that is due to cerebellar dysfunction (a broad differential that includes vascular, inflammatory, metabolic, hereditary, and degenerative etiologies); ALS due to progressive degeneration of both upper and lower motor neurons; MS due to demyelination of white matter; Parkinson plus syndromes such as multiple system atrophy and progressive supranuclear palsy; apraxic dysphonia; and Gilles de la Tourette syndrome. In ALS, oropharyngeal symptoms may predominate in up to 25% of patients, making it a disorder that may initially present to the ENT surgeon. In MS, ENT symptoms such as vertigo, dysphagia, slurred speech, and trigeminal neuritis are common, occurring in upward of 50% of patients. Thus, MS may also present initially to the ENT surgeon. MS often affects young individuals aged 20 to 50. Vocal spasticity is not uncommon, and thus MS must be considered in the voice clinic in young people with unexplained dysphonia.

Evaluation and Management of Neurological Voice Disorders

The key to evaluation and management of neurological voice disorders is the development of a multidisciplinary team. The speech-language therapist and the ENT surgeon will be at the heart of the team together with the neurologist and the neurophysiologist. It is helpful to have a good relationship with the psychologist and psychiatrist as well. A careful case history emphasizing the psychosocial aspects is fundamental to diagnosis as well as successful management. Woodson⁴⁹ notes that issues such as vocal fatigue, pain when speaking, increased effort relating to speech, glottic tightness, pitch breaks, and tremor should all be looked at. Nasendoscopic visualization and recording with playback to the patient is the gold standard.^50–52 Assessment by the speech-language therapist with a trial of voice therapy generally occurs, and all patients are referred for baseline neurological evaluation.

The speech-language therapist has a crucial contribution to differential diagnosis.⁵⁰ His/her role includes perceptual and instrumental assessment that are critical to measurement of baseline status and outcome; patient counseling in all cases; and voice therapy in some cases.

A thorough voice history including onset and progressiveness is necessary. Typical vocal signs and symptoms to watch out for include quality issues such as hoarseness, harshness, and strain/strangle; vocal effort issues such as vocal fatigue and breathy, reduced range; and pitch issues such as pitch breaks, monopitch, and inappropriately high pitch. Tremor and nasality are also common. Speech issues such as unclear, slurred, or unintelligible character are also important. During the history taking, the physician must also look for swallowing and respiratory issues. While many of these issues may occur in functional voice disorders, pronounced vocal fatigue after mild voice use may represent a disorder of the neuromuscular junction, a jerky tight tremulous voice may be due to defective motor control, and vocal weakness with incomplete adduction of the glottis may be secondary to paresis. Woodson49 reminds us that isolated hoarseness is not usually the only sign of a neurological disorder, while hoarseness and dysphagia likely point to a neurological basis.

The voice (and speech) examination begins with the ears to look for evidence of tympanic membrane retractions, glue ear, or other signs of eustachian tube dysfunction. It is also important to make certain that hearing loss is not an issue as it may impact on the way the individual hears their own or others’ voices.

The nasal examination includes a search for allergic rhinitis (as it is identifiable in over 20% of Europeans and can affect the nasality of the voice). Of great importance is to evaluate the closure of the soft palate against Passavant ridge in the postnasal space. Velopharyngeal insufficiency is a significant feature of many neurological disorders and will allow for air (and also fluid) to escape up into the nasal passages. Sounds such as “ah” and “ee” and high-pressure sentences such as “Suzy stayed all summer” will assist in this investigation. During speech, it is useful to pinch and release the nostrils.^11,53

Examination of the mouth is crucial. The character of the tongue should be looked for; for example, is it scalloped, discolored, or furred? Movement of the tongue is an important feature of the examination. Fasciculations may be an early feature of motor neuron disorders. And subtle or gross movement disorders on protrusion, retraction, and side-to-side tongue movement may be seen in several movement disorders as well as stroke. The integrity of cranial nerve XII should be assessed. Similarly, diminished voluntary soft palatal movement can be a prominent feature of disorders such as Parkinson plus syndromes and stroke. Palatal movement and sensory assessment are important to study as part of the evaluation of the integrity of the cranial nerves IX and X. During the mouth examination, the articulators should be assessed. Dysarthria is an extremely common symptom of stroke and the articulators can be checked with such sounds as/pa/,/ta/,/ga/. The fine control of the tongue against the hard palate, teeth, and lips should be checked.¹¹

Flexible nasendoscopy is, perhaps, the crucial evaluation parameter.⁵² During this examination, velopharyngeal insufficiency can be assessed for (as noted). The larynx and pharynx can then be examined in some detail. Observation of the larynx at rest for a full 2 minutes can provide important information about spontaneous movement of the larynx or other respiratory features and should be part of this examination. The larynx should be examined for clonic activity, tremor, supraglottic hyperfunction, paradoxical laryngeal movements, and hyperfunction or hypofunction at the level of the true glottis.

As a part of this examination, connected speech should be assayed as well as individual vowel sounds. The patient should be encouraged to repeatedly and rapidly sniff in (posterior cricoarytenoid muscle) alternating with a short/e/to assess for vocal fatigue (a feature of myasthenia gravis).

Immobility of one or both vocal folds can be noted and the appearance of flaccidity looked for. A rotated larynx with scissoring of the arytenoids is an indication of possible superior laryngeal nerve palsy. Pooling in one piriform together with one vocal fold held in the “cadaveric” position may be indicative of a high vagal lesion. We use a vocal protocol for assessment of laryngeal dystonia (see Box 19.1). It is useful for assessment of all patients with suspected neurological disorders affecting the larynx. It must be recalled that identification of an immobile vocal fold on nasendoscopy cannot definitively differentiate paralysis from cricoarytenoid fixation. EMG is required for this.

Stroboscopy is advisable in all of these patients.⁵⁴ Assessment should be formalized using criteria such as described by Ford and Bless.⁵⁵ Features such as symmetry, synchrony, mucosal wave aberrations, and closure patterns are best assessed using stroboscopy (at the time of this writing videokymography and high-speed photography do not have routine application in our clinic but may in the future).

In the clinic at the first assessment, a general neurological evaluation should be systemically performed. It should include (at least) complete assessment of cranial nerve function; assessment of cerebellar coordination (gait, finger–nose and heel–shin tests, etc.); assessment of Parkinson disease (resting tremor, rigid tone, and bradykinesia); assessment for upper motor neuron signs (spastic increase in tone on passive movements, brisk reflexes, and upgoing plantar responses); chorea and other movements (unintended tremor of the head or extremities, spontaneous facial movements, etc.).¹¹

Box 19.1 Laryngoscopy Protocol

Stay updated, free articles. Join our Telegram channel

Join