10.1.1 Symptoms

When patients have a paralysis or paresis of the vagus nerve or its branches, the resulting dysphonia typically results in a breathy voice. Patients frequently complain of having difficulty with projection or trouble talking over ambient noise. Phonatory dyspnea, vocal fatigue, and a higher pitched voice or paralytic falsetto are also common. They may note a weakened cough or even dyspnea on exertion due to the loss of autopositive end-expiratory pressure. They frequently will have coinciding dysphagia complaints as well, ranging from aspiration with liquids to dysphagia with all consistencies to nasal regurgitation. ^{3 , 4}

10.1.2 History

The initial otolaryngology assessment should start with a thorough history. Was the dysphonia sudden or gradual? Did it start after a recent surgery? The most common etiology for vocal fold paralysis is iatrogenic, followed by malignancy and idiopathic. ^{5 , 6 , 7} A history of any other neurologic symptoms should also be elucidated. Any associated symptoms such as those previously discussed should also be reviewed.

10.1.3 Examination

Next, examine the voice. The clinician should closely listen to the quality of the patient’s voice. It should be assessed for any breathiness or evidence of paralytic falsetto. As the patient slowly ascends in pitch by lengthening the vocal folds and reducing their flaccidity, they will demonstrate less breathiness, which is why some patients adopt the paralytic falsetto. When asked to project, patients with paralysis will frequently demonstrate a diplophonic voice (more than one tone is produced at the same time) that may not be present at quieter conversational volumes. The range is also frequently diminished, most commonly at the upper frequencies. The cough is frequently weak and the maximum phonation time is typically less than 10 seconds. ⁴ Occasionally, when the SLN is involved, innervation to the cricothyroid muscle is affected and the patient may note a lower pitch or monotone voice. ³

A complete head and neck examination should also be performed, particularly looking for any neck masses or associated cranial neuropathies. With a high vagal lesion that is more proximal along the course of the vagus before it begins to branch, the palate may be weak with a decreased palatal rise demonstrated on the paralyzed side as the pharyngeal branch of the vagus will be involved. Indirect mirror laryngoscopy can also be used to assess for the general mobility of the vocal folds or evidence of pooling of secretions.

Endoscopy ($115), whether flexible or rigid, can provide further information. ¹ In addition to evaluating gross movement, a more thorough evaluation of vertical and horizontal movements as well as evidence of vocal fold bowing or atrophy can be seen. With flexible laryngoscopy, the palatal movement and velopharyngeal seal can also be investigated as well as the lateral pharyngeal wall movement. By having the patient hold a loud and high pitch, normal pharyngeal contracture should be evident along the lateral and posterior wall. However, when one side of the pharynx is paralyzed as you would expect with a higher vagal lesion, the muscular bulging is lacking and the posterior midline raphe will frequently be pulled toward the nonparalyzed side. ⁴ In addition, secretions pooling within a unilateral pyriform also indicates pharyngeal weakness on that side. ▶Fig. 10.1 is an example of these findings; in the background, you can see an atrophied right vocal fold with bowing as well as secretions within the pyriform sinus, suggestive of a high vagal lesion. During the assessment, the tip of the scope can be used to test the laryngeal sensation. If sensation is absent, this points toward SLN involvement, which helps further localize the lesion.

When examining the overall appearance of the vocal folds, note the resting position as well as the contour. Muscle atrophy may cause bowing of the vocal fold, flaccidity, or an enlarged ventricle. ▶Fig. 10.2 demonstrates this in a patient with a left vocal fold paralysis. The left vocal fold is atrophied and bowed with resulting ventricle enlargement. With the addition of stroboscopy ($180), you can carefully evaluate glottic closure on phonation or mucosal wave asymmetry; however, when the immobile vocal fold is lateralized, it may be difficult to capture a signal. ¹ Flaccidity can also be evaluated by having the patient phonate at a low pitch; at a lower pitch, the vocal fold will begin to display lateral buckling and aperiodicity, which will resolve as the patient moves to a higher frequency. It is important to also note any vertical height mismatch that may be present as determined by the paralyzed vocal fold position and arytenoid rotation. While any position is possible, frequently the paralyzed vocal fold is shortened with anterior displacement of the arytenoid. ⁸

If a patient has had synkinetic reinnervation or a paresis, the examination findings are more subtle. Symptoms are similar to those found with complete paralysis and rarely associated with swallowing issues. On endoscopy, there typically appears to be hypomobility or bowing of the vocal fold. During the examination, it may be helpful to have the patient perform repeated tasks in hopes of fatiguing the pathologic side and making a vocal fold lag more noticeable. ³ Occasionally, the only examination finding may be asymmetry in vocal fold tension and supraglottic hyperfunction.

The SLN may also be involved, in either its sensory or its motor components. The sensory components are for supraglottic sensation and the motor function provides the innervation for the cricothyroid muscle. When the SLN is involved, the patient will frequently note a lowered pitch and more monotone speech. In addition, it can also cause vocal fatigue and breathiness. Due to the sensory component, patients may also experience choking and throat clearing. On examination, as the patient phonates at higher pitches, the posterior commissure will frequently rotate to the weaker side. With the flaccidity of the paralyzed side, this will cause it to appear shortened with bowing. ³

10.2.1 Imaging

Once paralysis is identified, an attempt to ascertain the etiology should be made. As malignancy remains one of the most common etiologies, this requires imaging along the course of the paralyzed nerve (from the skull base to the level of the xiphoid process). The etiology will also affect the treatment, as inflammation and compression may cause transient paralysis, while trauma or malignancy may be more permanent. The most commonly used imaging includes chest X-ray (CXR), ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI).

Optimal imaging is debatable and not well supported by evidence. In a survey of members of the American Broncho-Esophagological Association (ABEA), most respondents (70–73%) stated that a CXR or neck/chest CT was always or often needed in paralysis workup. However, the majority also said that MRI was only sometimes required. Ultrasound usage was not surveyed. ⁹

One study looked at contrast-enhanced head CT ($165), neck CT ($205), and chest CT ($200), for every patient who had an unknown clinical etiology (or no recent head, neck, or chest surgery that correlated with the onset). ¹ They looked at both paralysis and paresis. Routine imaging revealed lesions in 21% of the patients, with thyroid abnormalities being the most common finding. Ultimately, these lesions were thought to have contributed to the immobility in only 6% of the cases. ¹⁰ Other studies have shown that CT can identify the cause in 23.5% of patients with an unknown clinical etiology, with lung disease and thyroid disease being the most common causes. ¹¹ Another study showed that a chest CT identified the responsible lesion in 30.9%, the neck CT in 24.5%, and the brain CT in 14.8% of patients. ¹² The utility of CT imaging for paresis is much lower, with studies indicating yields ranging from 0 to 2.9%. ^{13 , 14}

In comparing CT to CXR ($30), CXR was diagnostic in 59% of the chest lesions and 80% of the cardiovascular lesions. As expected, CXR did not identify mediastinal disease well or lesions involving the skull base or neck. However, all of the chest lesions noted on CXR required contrast-enhanced CT for further evaluation and staging. Since a negative CXR does not exclude a malignancy and positive CXR still requires a CT for workup, it appears to have low utility. ¹¹

In comparing neck ultrasound ($120) to CT results, ultrasound was able to identify 100% of the cervical lesions and 12% of the chest lesions. ^{1 , 11} Another study showed comparable identification rates between a Neck CT and neck ultrasound, 24.5 and 26.2% respectively. ¹²

There are few studies regarding MRI. Brain, neck, and chest MRI cost $235, $325, and $465, respectively. ¹ In general, they seem to highlight that MRI is best for more “proximal” lesions as they provide better visualization of the skull base. ¹⁵ However, they have a high false-positive rate, especially in low-suspicion cases. ¹⁶

10.2.2 Serology

Occasionally, serology may be used in the workup. In the survey of ABEA members, most people stated that lab tests could be used; however, 80% stated these should be used occasionally or rarely. ⁹ The most commonly ordered tests were Lyme titer ($20), rheumatoid factor ($7), erythrocyte sedimentation rate ($5), and antinuclear antibody ($15). ¹ Unfortunately, the evidence for these tests is weak, with most of the articles being case reports. The only cross-sectional study that appeared to show any relationship between vocal fold paralysis and a systemic disease found that it was more common in diabetics, 0.44% in nondiabetics and 4 to 5.6% in diabetics. ¹⁷ A blood glucose would cost $5 and a hemoglobin A1c costs roughly $10. ¹ It is generally felt that given the low likelihood of these serologies being positive and the lack of evidence, they should only be ordered should there be suspicion for a particular disease.

10.2.3 Laryngeal Electromyography

Laryngeal electromyography (LEMG) may be performed by the otolaryngologist in conjunction with a neurophysiologist, to confirm vocal fold paralysis, postulate on the site of the lesion, and offer a view on recovery prognosis. According to a recent consensus statement, the optimal time to perform it is between 4 weeks and 6 months after initial insult. ¹⁸ Usually, the cricothyroid and thyroarytenoid muscles are tested to help determine SLN and RLN involvement. If both nerves are involved, this suggests a lesion proximal to their branching or a vagal injury.

In LEMG ($150), the change in the negative resting potential of the muscle cells is what produces the electrical activity. The basic component studied is the motor unit action potential (MUAP), which is the electrical summation of all muscle fiber potentials that are innervated by a single motor neuron and initially are bi- or triphasic waves. Immediately after a complete nerve injury, there is electrical silence at rest as well as with attempted movement. However, if the injury is incomplete, some fibers are still intact causing an MUAP with decreased amplitude. As reinnervation occurs, low-amplitude polyphasic MUAPs appear due to disordered reinnervation and muscle fiber atrophy with the amplitude increasing with time. ¹⁹ ▶Fig. 10.3 is an example of LEMG findings for a normally mobile thyroarytenoid (▶Fig. 10.3a) versus one with paralysis (▶Fig. 10.3b). Note that normally on contraction, there are multiple motor units firing and that a single MUAP can no longer be distinguished as the baseline is obscured (▶Fig. 10.3a). This is contrasted to (▶Fig. 10.3b) where there is a less robust electrical response and a clearly distinguishable baseline.

When the needle is initially inserted, a few fibers will depolarize before quieting down. If this activity is prolonged, this could indicate muscle instability or persistent denervation. At rest, muscle fibers usually have some spontaneous activity and negative deflections that do not propagate but if they are persistently firing and forming spike fibrillations or positive sharp waves, this is suggestive of denervation. Frequently, this can be seen starting 3 weeks after the initial injury, although some studies have indicated that this may happen even sooner. ¹⁹

In LEMG, findings of positive sharp waves, polyphasic MUAPs, or fibrillations are evidence of neurologic impairment. However, when it is paresis as opposed to paralysis, the signal of the remaining muscle activity usually obscures these findings. What may be seen with paresis instead is decreased recruitment, a finding that can also be mimicked by improper needle placement or incomplete muscle contraction. ²⁰ Also, striated muscle will still achieve a maximal interference pattern at only 30% of its maximum contraction because multiple motor units are firing at a high frequency and obscuring the features of a single MUAP, meaning paresis may not be even identified on EMG. With this in mind, LEMG may not be more accurate for diagnosing paresis than physical examination. In fact, a recent survey of laryngologists showed that most of them, 89%, relied on examination alone. The laryngoscopic findings that they thought to have the highest predictive value were motion anomalies such as sluggish motion and decreased tone. ²¹ However, it has also been shown that there is poor inter-rater agreement when presented with examinations suggestive of paresis, both on the diagnosis itself and on the lateralitiy. ²² When trying to correlate stroboscopy findings with LEMG-confirmed paresis, the examination findings that had the strongest correlation appear to be ipsilateral axis deviation, shorter vocal fold, thinner vocal fold, vocal fold bowing, reduced kinesis, and phase lag. ²³

There are some studies that suggest that LEMG can be used to assess the potential for recovery after paralysis. Criteria that are typically used to define “excellent prognosis” include no fibrillations or positive sharp waves with good motor recruitment. However, even using these criteria there is varied success, with studies indicating recovery in these patients anywhere from 13 to 90%. ^{24 , 25 , 26 , 27 , 28 , 29 , 30} A recent meta-analysis did reveal that the presence of MUAPs increased the likelihood of recovery by 53% over their absence, while there was insufficient evidence for the usefulness of fibrillation potentials and sharp waves. The absence of electrical synkinesis was also found to have a positive predictive value for recovery of 68% and a sensitivity of 93%. ¹⁸