Not paralysis, but dystonia causes stridor in multiple system atrophy

Materials and methods.

We examined 10 patients with probable MSA1 (4 patients with the cerebellar, 1 with the parkinsonian, and 5 with the mixed form), 4 women and 6 men (mean age 62.1 years [4.7]; mean duration of symptoms 5.7 years [2.7]); 7 patients had stridor, 6 had dysphagia, 4 patients had both, and 1 patient was asymptomatic for both.

The first seven patients examined were selected from our clinics because they had stridor. In addition, three patients with probable MSA, but without stridor, then were studied.

Stridor was defined as “a harsh strained inspiratory sound with higher pitch than snoring.” Severity of stridor was evaluated clinically and with the aid of a bed partner mini-questionnaire. It was classified as mild (+) when present rarely or sometimes only during sleep; moderate (++) when present often or always only during sleep; or severe (+++) when present often or always during sleep and wakefulness.

Dysphagia was defined as difficulty in swallowing. Severity of dysphagia was assessed based on the patient’s subjective complaint or problem. It was classified as mild (+) when the complaint was a feeling of something getting stuck while swallowing, or of coughing during or after swallowing; moderate (++) when patients had to cut food finely, avoiding certain difficult-to-particulate foods, or to swallow liquids in small sips; or severe (+++) when serious nutritional problems were present with need of a feeding tube.

Patients’ characteristics are summarized in the table.

Informed consent was obtained from all patients.

All patients were investigated by fiberoptic laryngoscopy to evaluate vocal cord activity and morphologic features. Electromyographic investigation of laryngeal and pharyngeal muscles was performed using concentric needle electrode. Simultaneous recording from the thyroarytenoid muscle (vocal cord adductor agonist) and the posterior cricoarytenoid muscle (vocal cord abductor) were performed in all patients during quiet breathing. Subjects were examined in supine position with their head elevated at a 30° angle. Lidocaine spray solution (4%) was applied over the skin of the cricothyroid space. EMG signals were band-pass filtered between 10 Hz and 5 kHz. The thyroarytenoid muscle was examined by inserting the needle electrode in the cricothyroid space through the cricothyroid membrane, 2 mm above the upper edge of the cricoid arch. The needle then was directed posteriorly, superiorly, and laterally at an angle of 45° to reach the muscle bundles of the thyroarytenoid muscle. For the posterior cricoarytenoid muscle, a bent needle electrode was inserted at the cross point between the posterior border of the thyroid cartilage and the horizontal line of the cricoid space, moving medially and posteriorly around the cricoid cartilage.7 EMG investigation of the cricopharyngeal muscle was performed in eight patients; the needle electrode was inserted through the skin at the level of the cricoid cartilage, 1.5 cm lateral to its palpable lateral border, in the posteromedial direction.8 At rest, the cricopharyngeal muscle presents a tonic EMG activity in relation to its function as upper esophageal sphincter; such activity typically disappears (EMG silence) during the hypopharyngeal phase of deglutition. Patients were examined at rest and when swallowing 3 mL of water.

Amplitude and duration of motor unit action potentials (MUAP) of the patients were compared with our laboratory reference values obtained by EMG testing of 22 normal subjects (age range 25–75 years, mean age 55 years): for thyroarytenoid muscle, MUAP amplitude was 100–1100 μV, and MUAP duration was 3.2–6.3 milliseconds; for posterior cricoarytenoid muscle, MUAP amplitude was 105–980 μV, and MUAP duration was 3.0–6.3 milliseconds; and for cricopharyngeal muscle, MUAP amplitude was 110–1300 μV, and MUAP duration was 3.5–7.8 milliseconds.

In four patients with severe stridor, botulinum toxin (Botox, 10 IU) was injected into the right thyroarytenoid muscle, and its effect was evaluated 1 month after inoculation.

Results.

In the seven patients with stridor, laryngoscopy showed vocal cords bilaterally in a paramedian position and abnormal vocal cord movements. Three patients without stridor had normally abducted vocal cords with normal mobility (see the table).

In all muscles examined in our patients, the EMG did not show any evidence of spontaneous activity from denervation (i.e., fibrillation potentials, complex repetitive discharges, positive sharp waves), and both amplitude and duration of MUAP were in the normal range. In the seven patients with stridor a rapidly recruiting, prolonged EMG tonic activity (5–15 seconds) alternating with short periods (0.2–0.5 seconds) of electric silence was simultaneously present in the thyroarytenoid and the posterior cricoarytenoid muscle during quiet breathing. In two of the three patients without stridor, laryngeal EMG was normal; in the third patient, EMG showed bursts of activity in the thyroarytenoid muscle only during inspiration.

In four of the five patients with dysphagia examined, EMG activity of the cricopharyngeal muscle was abnormally persistent and did not show the normal electric silence during the hypopharyngeal phase of deglutition; electric silence was present in the fifth patient. Three patients without dysphagia had the normal EMG silence. One patient with dysphagia and one without declined the EMG study of the cricopharyngeal muscle.

Three of the four patients in whom botulinum toxin was injected into the thyroarytenoid muscle experienced net improvement of the stridor 1 month after inoculation. This was confirmed by both laryngoscopy (right vocal cord abducted) and on the EMG (reduction of persistent tonic activity in the right thyroarytenoid muscle) (figure). In one patient, response to botulinum toxin was unsatisfactory, and tracheostomy was required.

• Download figure
• Open in new tab
• Download powerpoint

Figure. Laryngoscopy (top) and electromyogram (EMG) activity (bottom) before and after injection of botulinum toxin. Patient 2 before (left) and after (right) injection of botulinum toxin into the right thyroarytenoid muscle (arrow). Simultaneous EMG recording was performed from the right posterior cricoarytenoid muscle (top) and the right thyroarytenoid muscle (bottom). One month after injection, laryngoscopy showed the right vocal cord abducted; EMG activity of the right thyroarytenoid muscle was markedly reduced. The slight reduction of EMG activity observed in the right posterior cricoarytenoid muscle could be explained by either a “balance” mechanism (reduced activity of the antagonist muscle) or diffusion of botulinum toxin from the site of inoculation.

Discussion.

In this study of 10 patients with probable MSA, EMG investigation and response to botulinum toxin injection confirmed that the cause of stridor in MSA is hyperactivity of vocal cord adductors, not paralysis of vocal cord abductors.

These sustained contractions of vocal cord adductors and cricopharyngeal muscle causing their abnormal postures and their altered motility (i.e., stridor and loss of deglutitive inhibition) can be referred to as dystonia, affecting the musculature at the laryngeal–pharyngeal level.9

Our finding also is supported by the following observations. First, neuropathologic studies in MSA do not show neuronal loss or alterations of micromorphologic features in the nucleus ambiguous3,10⇓ (the nucleus of origin of the motor fibers that supply the striated muscles of larynx and pharynx). Second, stridor in MSA presents as a transient, short-lasting phenomenon, often exacerbated by emotion; this is more in keeping with a dystonic disorder that typically fluctuates than with paralysis. Third, the EMG study showed no evidence of neurogenic degeneration in any of the laryngeal–pharyngeal muscles examined.

The abnormal EMG pattern of vocal cord adductors observed in one patient without stridor (bursts of EMG activity during inspiration) could be interpreted as a preclinical alteration of adductor motility; however, it is possible that mild stridor was not detected by the bed partner. Conversely, the normal EMG swallowing pattern in one patient with mild dysphagia could be explained either by the volume of water used for testing, too small to disclose the symptom, or with the presence of coexisting mechanisms responsible for dysphagia in MSA.

Laryngoscopy cannot provide a complete functional evaluation of vocal cords: the observed paramedian position can either be the result of vocal cord abductor paralysis or of hyperactivity of vocal cord adductors in conjunction with functional integrity of abductors.

In this study, laryngeal EMG and the response to botulinum toxin confirmed the latter condition.

The relative atrophy of the vocal cord abductors observed in a neuropathologic study of MSA patients2 could be explained by an imbalance between agonist and antagonist caused by the hyperactive adductors.

Evidence of the dystonic nature of stridor and dysphagia in MSA has important repercussions on the treatment choice.

In this study, botulinum toxin was injected unilaterally, mainly to confirm our clinical, laryngoscopic, and electrophysiologic observations. Bilateral injections of adequate doses of botulinum toxin into the thyroarytenoid muscles might improve efficacy.