Introduction
Laryngeal dystonia (LD) (also known as spasmodic dysphonia) is a chronic neurologic disorder of central motor processing with action-induced, task-specific focal dystonia. It most often results in a disturbance in the production of connected speech. Females are preferentially affected, and the average age of onset is 45 years. Patients present with irregular voice breaks, effortful phonation, and breathiness that are thought to start gradually and then progress slowly before becoming chronic. In a subset of patients, the disease course may not be progressive and usually waxes and wanes. Clinically, there are 2 major types of LD: adductor laryngeal dystonia (ADLD) and abductor laryngeal dystonia (ABLD). ADLD is characterized by a harsh, strain-strangled voice with prominent breaks on vowels in speech, diminished volume, and a monotonal pitch. Patients with ABLD present with a breathy, effortful voice and abrupt voice breaks on ongoing phonation. With severe spasms, patients may become aphonic or locked in a whisper. Compensatory behaviors may mask the patient’s true voice pattern. Other disease variants, such as singer’s dystonia and adductor respiratory dystonia, have also been described. ,
The diagnosis of LD can be difficult due to the lack of a scientific consensus on diagnostic criteria and the fact that other voice disorders may present similarly. In addition, there are no pathognomonic features of the history or examination. Ludlow et al. recommended a 3-tiered approach to diagnosing LD, involving a questionnaire, followed by speech assessment and nasolaryngoscopy. This study highlighted the difficulties of gathering consensus on diagnosis among experts. Muscle tension dysphonia (MTD), a functional voice disturbance, and vocal tremor (VT) are the most important entities of the differential diagnosis. The hyperadduction of MTD is generally sustained and unlikely to be spasmodic, whereas the involuntary movement in VT is rhythmic and often involves pharyngeal muscles and strap muscles. In addition, motion abnormalities in LD are present only during specific speaking actions, whereas they are present more consistently in MTD and VT.
In many cases, the initial treatment for patients with suspected LD is voice therapy, even when voice characteristics suggest a neurogenic entity. Voice therapy is noninvasive and individualized, and the response to therapy can help support the diagnosis. Voice therapy can reduce compensatory behaviors that patients with LD often experience after botulinum toxin (Botox) treatment and provide strategies for side effects, including the initial breathy period. Therapeutic laryngeal Botox injections are offered to patients with a clinical history consistent with neurogenic dysphonia, a perceptual voice exam characterized by voice breaks, a laryngoscopic exam that demonstrates hyperspastic features, and a lack of response to voice therapy. Patients displaying other types of tremors, other neurologic findings, or a family history of neurologic disease should be referred to a neurologist for further evaluation.
Treatment
Treatment Alternatives to Botulinum Toxin
There is no specific pharmacologic treatment for LD. Oral medications used in dystonia, such as anticholinergics, benzodiazepines, and baclofen, provide minimal relief and have numerous side effects at the doses required to influence a patient’s voice. Some patients report symptom improvement with alcohol, which suggests that processes affected by alcohol might respond to pharmacologic intervention. Sodium oxybate (Xyrem, Jazz Pharmaceuticals, Dublin, Ireland), a GABAergic medication, may be a promising pharmacological agent for the treatment of ethanol-responsive LD and VT patients.
Voice therapy can be useful as an adjunctive treatment to botulinum neurotoxin (BoNT) injections by addressing the compensatory behaviors superimposed on LD. Still, it does not yield marked improvement by itself. Voice therapy is the first-line therapy for MTD and is helpful as a diagnostic maneuver in ambiguous cases.
The early surgical approaches to symptom control in LD involved performing a unilateral recurrent laryngeal nerve section. Subsequently, several surgical procedures have been developed for the treatment of LD, including recurrent laryngeal nerve avulsion, laryngeal framework surgery, partial and total laser thyroarytenoid muscle resection, implantable stimulators, and selective laryngeal adductor denervation-reinnervation using ansa cervicalis. The underlying principle is the same as with BoNT: denervation. Surgical treatments demonstrate short-term improvement, but a significant number of patients show recurrence of their symptoms after a period of time.
Botulinum Toxin Treatment
BoNT type A injections are the first-line treatment for LD. The first laryngeal Botox injection for ADLD was given in April 1984 by Blitzer et al. The standard treatment for ADLD is bilateral EMG-guided transcutaneous injections into the thyroarytenoid muscle. Initial doses vary widely amongst practitioners. , The effective dose is not proportional to body mass or dysphonia severity and varies considerably across patients, with an average dose of 0.9 U/0.1 mL per vocal fold. The surgeon can arrive at the correct amount only through experience with each case. For this reason, a record of dose and effect, both beneficial and adverse, is kept for each patient.
The strategy in the abductor group is to treat one posterior cricoarytenoid (PCA) muscle at a time. Excessive abductor weakness results in an inability to abduct the vocal folds during inspiration, potentially causing stridor and dyspnea. Treatment is started with 3.75 U/0.15 mL in the PCA. Laryngoscopy is often performed before injection to determine which side is more spastic.
Outpatient Techniques for Vocal Fold Injections
Several office-based techniques have been described for accessing the intrinsic laryngeal musculature. This chapter will focus on the EMG-guided percutaneous and trans-nasal approaches.
EMG-guided Percutaneous Injection Technique
BoNT is injected using a hollow-bore 27-gauge Teflon-coated needle attached to an EMG, functioning as a monopolar electrode. A standard dilution of 4.0 mL of sterile saline/100-U vial of Botox (Allergan, Inc., Irvine, CA, US) results in a concentration of 2.5 mouse units per 0.1 mL, which can be further diluted as necessary for each patient to deliver the appropriate dose in a typical volume of 0.1 mL per vocal fold. Smaller-volume injections are meant to prevent airway obstruction and unwanted diffusion of the drug.
Adductor Laryngeal Dystonia
The target for the treatment of ADLD is the thyroarytenoid muscle. It originates at the inner surface of the angle of the thyroid cartilage and cricothyroid ligament and inserts on the anterolateral surface of the ipsilateral arytenoid cartilage. Its proximity to the cricothyroid membrane allows for the injection of the muscle through the cricothyroid space.
To treat the thyroarytenoid muscle, the patient is placed in a supine position or beach chair position with the neck slightly extended. The thyroid and cricoid cartilages are identified by palpation, and the cricothyroid membrane is delineated. The needle is bent upwards 30–45 degrees; this is especially important when injecting the female larynx, as the shorter anterior-posterior distance requires a more acute angle of entry under the inferior border of the thyroid cartilage ( Fig. 43.1 ). ,
EMG-guided injection of the thyroarytenoid muscle for adductor laryngeal dystonia.
(From Sulica L, Blitzer A. Botulinum toxin treatment of spasmodic dysphonia. Oper Tech Otolaryngol Head Neck Surg . 2004;15(2):76–80.)
The EMG needle is transcutaneously inserted through the cricothyroid membrane, 2–3 mm from the midline of the side to be addressed. It is advanced superiorly and laterally toward the target thyroarytenoid muscle. Often, the needle enters the air column in the laryngeal lumen after traversing the cricothyroid membrane, producing a characteristic “buzz” on EMG. This tells the surgeon that the needle needs to be directed laterally. Crossing the endolaryngeal mucosa can irritate the patient, triggering a cough reflex. By piercing the cricothyroid membrane a few millimeters to the side of the midline, the surgeon may enter the thyroarytenoid muscle directly without first entering the airway. Once the needle is placed in an area that demonstrates crisp motor unit potentials, placement in the correct muscle is confirmed by asking the patient to phonate. Brisk recruitment in EMG confirms placement in the adductor muscles. The surgeon then aspirates to ensure that the needle has not pierced a blood vessel, and BoNT is then injected. ,
Abductor Laryngeal Dystonia
The PCA is the target muscle for the treatment of ABLD. It arises from the depressions of the posterior surface of the cricoid lamina and extends superolaterally to insert onto the posterior aspect of the ipsilateral muscular process of the arytenoid cartilage.
The PCA may be reached in one of two ways. Most commonly, the surgeon rotates the laryngeal framework laterally for direct access to the posterior aspect of the larynx from the lateral neck. The patient should be positioned with the neck and shoulders relaxed to allow for minimal muscle activity. The surgeon places his or her thumb at the posterior edge of the thyroid cartilage on the side to be injected and applies counterpressure with the other four fingers on the contralateral thyroid lamina. The entire larynx is then rotated to expose its posterior aspect ( Fig. 43.2 ). The vascular bundle is pushed posteriorly to make sure there is a plane of separation between the bundle and the thyroid lamina. The EMG needle is inserted along the lower half of the posterior edge of the thyroid cartilage, traversing the inferior constrictor, and advanced until it abuts the surface of the cricoid cartilage. It is then pulled back slightly under EMG guidance. The patient is asked to sniff, which yields maximum abduction. The EMG signal is observed for correct placement, and then the BoNT is injected in an area of brisk activity. This technique works best in patients with a thin neck and a high, mobile larynx. , ,
Retrocricoid EMG-guided injection of the posterior cricoarytenoid muscle for abductor laryngeal dystonia.
(From Sulica L, Blitzer A. Botulinum toxin treatment of spasmodic dysphonia. Oper Tech Otolaryngol Head Neck Surg . 2004;15(2):76–80.)
Alternatively, the needle may be inserted through the cricothyroid membrane in the midline, guided across the subglottic space (identified by the characteristic “buzz”), and advanced through the posterior lamina of the cricoid cartilage to one side or the other of the midline ( Fig. 43.3 ). An intratracheal injection of 1% lidocaine helps prevent coughing as the needle crosses the lumen of the larynx. This does not hinder EMG signaling as the target muscle is on the opposite side of the cricoid cartilage. Once through the cricoid cartilage, the first signal encountered on the far side represents the PCA muscle. Placement is confirmed by asking the patient to sniff, and then the BoNT is injected. This approach is most useful in young patients with soft cartilage. However, it is not recommended in older patients with extensive cartilage calcification, which makes this technique difficult. In addition, fragments of cartilage often plug the needle as it passes through the cricoid and may require substantial plunger pressure to clear. Using a syringe with a Luer lock is helpful to prevent the separation of the needle in this event. Because of potentially life-threatening dyspnea, only one side is treated at a time, waiting approximately 2 weeks for the peak effect to pass before injecting the contralateral side. , ,
Transcricoid EMG-guided injection of the posterior cricoarytenoid muscle for abductor laryngeal dystonia.
(From Sulica L, Blitzer A. Botulinum toxin treatment of spasmodic dysphonia. Oper Tech Otolaryngol Head Neck Surg . 2004;15(2):76–80.)
Trans-nasal Injection Technique
The technique for injecting BoNT into the thyroarytenoid muscles for ADLD through a needle placed via an operative channel of a flexible laryngoscope was first described by Rhew et al. Both a flexible laryngoscope with a working channel and a catheter needle are required for this approach.
Before the procedure, patients are premedicated with 1% phenylephrine sprayed into each nostril to decongest the nasal mucosa so that the passage of the flexible laryngoscope causes less discomfort. Topical anesthesia is then accomplished through several measures: (1) 4% lidocaine is sprayed into each nostril and the posterior pharynx and hypopharynx; (2) the patient is given 5 mL of 2% viscous lidocaine to gargle and spit; and (3) the laryngoscope is passed through one nostril down to the level of the endolarynx, and 5 mL of topical xylocaine is dripped into the larynx using a 5 mL syringe with an attached cannula. Attachment of the cannula facilitates injection of xylocaine directly into the lumen of the side channel, rather than at the opening. , Alternate topicalization methods include a direct transtracheal injection of 4% lidocaine into the trachea or nebulized lidocaine treatment.
The injection needle is primed with Botox (Allergan, Inc., Irvine, CA, US) at a dilution of 2.5 units per 0.1 mL to eliminate dead space. The side channel is loaded with the catheter needle until 1 mm of the tip is visible beyond the end of the flexible laryngoscope. After 5 minutes, the surgeon performs a flexible laryngoscopy and positions the end of the scope just above the site to be injected (a mid-cordal injection site, lateral to the arcuate line, is preferred because the bulk of the thyroarytenoid muscle lies there). The assistant then passes the needle into the thyroarytenoid muscle under direct vision and injects the appropriate dose of BoNT, observing for any leakage from the puncture site. Such leakage is indicative of too shallow an insertion and can be corrected by inserting the scope a few millimeters lateral to the initial site. If the opposite vocal fold is to be injected, the tip of the scope is then positioned over the second site, and the procedure is repeated.
The advantage of this technique is that no EMG is required, and the false vocal folds can be targeted for treatment in patients with significant supraglottic compression. However, it also has several disadvantages. It takes longer than the percutaneous method, representing the time required for the topical anesthetic to become fully effective. Some patients are unable to tolerate flexible laryngoscopy even with topical anesthesia. In addition, injection through a flexible laryngoscope entails a waste of BoNT to fill the needle tubing and makes precision difficult to control. Lastly, this technique does not allow for EMG confirmation of needle placement, which allows for controlled administration of treatment into more actively contracting regions of the muscle.
Outcomes
Patients with ADLD treated with BoNT have reported speech improvement from approximately 52% at baseline to 90% with treatment, with an average duration of benefit of approximately 15 weeks. ABLD patients have also reported benefit, although more modest (an improvement from approximately 55% to 67%) and of shorter duration (10.5 weeks), probably because treatment is technically more difficult and is limited by respiratory complications should both PCA muscles be too aggressively denervated. Because the therapeutic effect of BoNT is temporary, patients require repeated BoNT injections every 3–6 months to sustain the therapeutic effect. Additionally, due to the potential airway issues limiting the ability to administer simultaneous bilateral injections, greater than 30% of patients with ABLD are also on low-dose oral agents.
Complications
The most commonly reported side effects of the adductor injections include a mild, transient breathy voice (reported by 25% of patients); mild, transient cough on drinking fluids (reported by 10% of patients); and local pain, bruising, or itch (reported by <1% of patients). In contrast, abductor injections are associated with mild exertional wheezing (reported by 2% of individuals) and mild transient dysphagia to solids (reported by 6% of individuals).
Failure rates of BoNT injections for LD range between 6% and 29%. , An early study examined the rates of treatment failure using different injection methods and different formulations of BoNT A and found that the failure rates did not differ with method or formulation. In a more recent study by Zhao K et al., the failure rate was 12% and was associated with patients’ first injection with a physician, professional voice user, and lack of breathiness.
Although rare, patients can develop secondary resistance to BoNT type A therapy, possibly due to the development of neutralizing antibodies. Most cases of secondary resistance to BoNT type A are designated after there has been a failure to respond to an appropriate or increased dose of BoNT, although this may represent disease progression and not necessarily antibody formation. The incidence of secondary resistance to BoNT type A in laryngeal disorders is not well studied. Park et al. reported their incidence to be 8.5% in a report sampling 71 patients over 5 years. BoNT type B is a therapeutic option for patients who are nonresponders or poor responders to BoNT type A.
