25

Spasmodic Dysphonia and Muscle Tension Dysphonia

Gayle E. Woodson

Spasmodic dysphonia (SD) and muscle tension dysphonia (MTD) are voice disorders that disrupt speech and can sound very similar.^1,2 The clinical description of vocal quality in patients with adductor SD is “strained and strangled.” Patients with MTD can also speak with a voice that is “strained and strangled,” and even experienced and expert listeners find it difficult to differentiate between the sounds of these voice disorders. But the etiology and management of these problems is quite different. SD is a neurologic disorder.^3,4 MTD is believed to result from habitual misuse of the voice. What the two disorders do have in common is that the voice disorder is not due to a structural abnormality in the larynx, although MTD can lead to secondary lesions. Differentiation between SD and MTD can be accomplished by listening to the voice and examining the larynx with continuous light. Videostroboscopy does not usually play a role in this differentiation. However, in patients suspected of having MTD, videostroboscopy has a more important role. Presumed MTD patients who do not respond to voice therapy may actually have compensatory hyperfunction due to a mucosal or submucosal lesion that is not apparent on examination with continuous light.

For many years, SD was referred to as “spastic dysphonia.” It was first described in 1871 by Traube, who reported that the sound of the voice was like that of a person trying to speak “whilst choking.”⁵ He believed that it was a psychiatric disorder. A psychiatric cause was accepted for many years, because no organic cause could be identified, and patients frequently reported that the onset of symptoms was during a time of stress. Because it is a debilitating disorder, most patients become discouraged, socially isolated, and even depressed. SD does not respond to any medical therapy, and speech therapy is only occasionally effective. However, in 1975, Dr. Herbert Dedo reported that surgically transecting the recurrent laryngeal nerve dramatically improved the voice in many patients with severe SD.⁶ The efficacy of this surgical intervention provided strong evidence for an organic cause of the disease. Subsequently, injection of botulinum toxin into laryngeal muscles was shown to be effective in reducing the muscle spasm of SD, without totally denervating laryngeal muscles.^7–9 Psychological testing in patients before and after effective treatment of SD has documented improvement in mood and anxiety, indicating that the emotional disturbances observed in patients with SD are the result of the disorder rather than its cause.¹⁰

SD is now recognized to be a neurologic disorder. Specifically, it is a focal dystonia of the larynx, characterized by involuntary spasms of the intrinsic laryngeal muscles during speech. There are two basic forms of the disorder, adductor SD (AdSD) and abductor (AbSD). In AdSD, the most common form, the voice is strained due to spasmodic closure of the glottis.11 In AbSD, the spasms abduct the vocal folds, opening the glottis, so that the voice is breathy.¹² Some patients have both adductor and abductor spasms. Speech is very effortful for patients with SD and they are clearly uncomfortable; however, there is no pain. It does not hurt to talk.

The muscle spasms in AdSD cause episodic and transient hyperclosure of the glottis, resulting in voice stoppage or pitch breaks. The occurrence of such voice breaks is the one feature that seems to be pathognomonic for AdSD, as these breaks have not been observed in patients with MTD. Adductor spasms are most prominent when the patient speaks words beginning with a vowel, such as “eggs” or “every.” The frequency of voice breaks can be assessed by listening or quantified by measuring events in a sample of an acoustic recording of the voice. The frequency of voice breaks has been shown to decrease significantly in patients with symptomatic improvement from botulinum toxin treatment.^8,9 But the voice in patients with SD is not only disrupted by the direct effects of these muscle spasms. Patients adopt varying strategies to cope with the spasm, and these behaviors can significantly influence speech. For example, some patients with AdSD will whisper. Others may adopt vocal hyperfunction in an attempt to stabilize the voice. This is quite similar to the vocal behavior observed in patients with MTD, with false vocal fold closure, anterior-posterior squeezing of the glottis, and excess tension in cervical strap muscles. Less commonly, patients with AdSD learn to phonate on inspiration.

In AbSD, the voice is disrupted by sudden abductor spasms or delays in adduction at the onset of phonation, resulting in episodic or persistent breathy voice. These patients usually have breathiness during production of sustained vowels. The spasms are most prominent in words or syllables that begin with a voiceless consonant followed by a vowel. For example, “pay” begins with a voiceless consonant, “p,” followed by a voiced vowel, “a.” Pronunciation of this word requires abduction of the vocal folds during the “p,” followed by rapid adduction of the vocal folds to produce the vowel. The vocal folds seem to “hang up” in the abducted position, so that the vowel is breathy, or even aphonic. As with AdSD patients, AbSD patients tend to adopt compensatory strategies that can be very similar to vocal behavior in MTSD.

The hyperfunctional voice in MTD is believed to involve excess activity in the extrinsic laryngeal muscles, including the cervical strap muscles.^1,2 There are no physiologic studies to support this assumption; however, physical examination in these patients usually reveals tight contraction and sometimes even tenderness in the cervical strap muscles. Additionally, it has been reported that many patients with MTD improve with speech therapy, specifically massage of the extrinsic laryngeal mucles.¹³ There have been no controlled studies to rule out a placebo effect of massage. However, MTD is generally observed to be responsive to speech therapy. This response is a key clinical feature that can distinguish MTD from SD. But careful evaluation is usually effective in identifying patients with SD.

Clinicians should listen to patients perform a variety of vocal tasks. A standard assessment protocol has been recommended by the Research Planning Workshop on Spasmodic Dysphonia that convened at the National Institute on Deafness and Other Communication Disorders (NIDCD) in June 2005.¹⁴ The voice should be assessed during conversational speech, sustained phonation, counting, whispering, and shouting, assessing vocal strain, detecting vocal tremor, and quantifying the frequency of voice breaks. Voice breaks and vocal strain should be absent during whispering and shouting, and the phonation is normal during laughter. Table 25.1 lists 10 standardized sentences that the workshop recommended for use in evaluating patients with SD.¹⁴ Ten of the sentences are more difficult for patients with AdSD, and 10 are more difficult for patient with AbSD. It is not necessary to use all 20 sentences, but the clinician should ask the patient to speak as many of these sentences as necessary to clarify the diagnosis. Patients with SD should produce at least three breaks per sentence when using a normal voice but no breaks when speaking in a whisper. To differentiate between AdSD and AbSD, the patient should be asked to alternately speak sentences from the two groups. Group 1 sentences include words that begin with a vowel, as vowel onset is particularly prone to elicit adductor spasm. Group 2 sentences include words and syllables that begin with unvoiced consonants, immediately followed by a vowel. The transition from an unvoiced consonant to phonation elicits abductor spasm in AbSD patients, interfering with the glottic closure that is required for phonation. Thus, patients with AbSD should have more breathy breaks when speaking sentences from list 2 than when speaking sentences from list 1.

Laryngeal behavior should be observed during a variety of phonatory and nonphonatory tasks. This can only be accomplished by flexible endoscopy, which permits observation of the larynx without interfering with its function during speech.15 Either a fiberoptic endoscope or a flexible endoscope with a distal chip camera can be used. The enlarged image on a video monitor is superior for detecting mucosal pathology and assessing motion. A video recording of the examination can be replayed repeatedly and viewed in slow motion to provide detailed analysis of movement. It is also very useful to save the recording for comparison with future evaluations.

On passing the endoscope from the nasal cavity to the pharynx, function of the soft palate should be assessed. Patients with SD may have an associated tremor of the larynx and pharynx that also affects the palate. If so, this tremor can be observed during phonation of “ee.” Observation of the palate is also important to exclude other neurologic impairments that could present with disrupted speech. The palate should elevate symmetrically and close snugly during swallow and phonation of “ee.” Abnormal palate motion indicates that there is some other neurologic disorder, not SD. The endoscope should then be advanced into the hypopharynx to look for lesions and to assess function. Pooling of secretions in the hypopharynx indicates possible sensory or motor impairment.16

The larynx should be observed during breathing to confirm normal vocal fold motion with respiration. The vocal folds may appear motionless during quiet breathing, but the degree of motion varies with effort. Therefore, the patient should be instructed to sniff and cough to induce a greater range of motion. The vocal folds should abduct briskly during a sniff. Alternating sniffing with brief phonation is a good way to demonstrate the range of motion of the vocal folds. A voluntary cough is very useful to establish the potential range of motion in the larynx. There are three phases of laryngeal motion during a cough, whether spontaneous or voluntary. First the glottis opens widely for deep inspiration and then closes for the second stage, the compressive phase, with expiratory force against the closed glottis. In the third phase, the glottis suddenly opens widely while expiratory force continues, resulting in a sudden outrush of air.

Observation of laryngeal motion during speech is very important to detect hyperfunctional posturing and spasmodic activity and to differentiate between adductor and abductor spasms. Patients with SD often have an associated tremor of the larynx and sometimes the pharynx as well. As a rule, MTD patients do not have tremor. The larynx should be observed during sustained phonation as well as in recitation of selected sentences from the lists of recommended standard sentences. In AdSD, the vocal folds are continuously or episodically closed tightly, with spasmodic increases in adduction, particularly during speaking of sentences in list 1. In patients with AbSD, sudden vocal fold abduction and/or sluggish adduction is best observed during transition from voiceless consonant to vowel. These transitions are frequent in speaking of sentences in list 2. The abductor spasms may be difficult to detect due to vertical motion of the larynx and movement of the epiglottis. Detection is facilitated by playing the recorded examination back in slow motion. Spasms are usually asymmetric in AbSD. Abduction is usually much greater on one side than on the other. Determination of the dominant side is important in planning treatment with botulinum toxin or medialization surgery.¹⁷

Hyperfunctional behavior during speech is observed in some patients with SD as well as in all patients with MTD. Laryngeal manifestations of hyperfunction include anterior-posterior compression of the larynx and supraglottic constriction, with adduction of the false vocal folds and aryepiglottic muscles. In SD, hyperfunction is believed to be compensatory for the underlying spasm, and there may be great variation in laryngeal movement. In contrast, the hyperfunctional behavior in MTD is much more continuous and consistent. Hyperfunction and hoarseness in MTD patients do not vary with pitch or between sentences in the two standard sentence lists. Voice therapy probing can be conducted during the laryngeal examination, taking the patient through maneuvers to relax the larynx. Improvement during such probe therapy is strong evidence for MTD.

Videostroboscopy

Stroboscopy is rarely useful in the diagnosis of SD, as the diagnosis is established on the basis of voice evaluation and laryngeal examination. Moreover, the normal sound of the voice during shouting and laughter is strong evidence against any structural defect in the vocal fold. However, it is conceivable that a patient with SD could have a coexistent mucosal lesion of the larynx and that the diagnosis could be more elusive. In such a case, stroboscopic evaluation would be indicated. But it is often difficult to obtain a satisfactory stroboscopic examination of the larynx in a patient with SD. This is because stroboscopic illumination must be flashing very near the same frequency of the vibration of vocal fold mucosa. This is generally accomplished by acoustic analysis of the voice signal to extract the fundamental frequency. In severe SD, or in any patient with severe hoarseness, the acoustic signal of the voice is so irregular that a fundamental frequency cannot be identified. In many patients with SD, there are intervals of fairly stable vibration between vocal breaks, but these intervals are often too short to provide a stable signal.

The diagnosis of SD and MTD is primarily based on clinical evaluation and standard laryngeal examination. Videostroboscopy is difficult to perform in patients with significant dysphonia because the frequency of vibration is too irregular to trigger the strobe. In cases of presumed MTD that do not respond to voice therapy, the possibility of structural pathology of the vocal fold must be considered, and stroboscopic examination is indicated. Alternate techniques of triggering the stroboscope can be helpful in these cases.

Stay updated, free articles. Join our Telegram channel

Join