CHAPTER 66 Medialization Thyroplasty
Since the first writing of this chapter, laryngeal phonosurgery has undergone evolution comprising a continuum of procedures designed to rehabilitate the dysfunctional larynx. Phonosurgical procedures may be classified into the following categories: (1) microlaryngeal procedures for excision of benign or malignant disease, (2) vocal fold injection for augmentation and medialization, (3) laryngeal framework surgery, (4) laryngeal reinnervation procedures, and (5) reconstructive and rehabilitative procedures after tumor resection. Laryngeal framework surgery has been further categorized by Isshiki and associates1 into the following four types of surgical procedures on the basis of the functional alteration of the vocal folds: medial displacement (type I), lateral displacement (type II), shortening or relaxation (type III), and elongation or tensioning procedures (type IV).2–5 This chapter focuses on laryngeal framework surgery and vocal fold injection in the context of rehabilitation of the paralyzed larynx while specifically addressing the management of glottal insufficiency from unilateral vocal fold motion impairment, vocal fold bowing, and soft tissue deficits.
The predominant focus of phonosurgical procedures has been rehabilitation of the paralyzed larynx. With a few exceptions, primary repair by end-to-end anastomosis after injury to the recurrent laryngeal nerve has been universally unsuccessful.6 Failure of primary repair is ascribed to a random process of axonal regeneration at the site of injury that results in the simultaneous contraction of antagonistic muscle groups, otherwise known as synkinesis.7,8 Alternative methods of reinnervation using ansa hypoglossus nerve–muscle pedicle implants into the posterior cricoarytenoid muscle and phrenic nerve–recurrent laryngeal nerve anastomosis for bilateral vocal fold paralysis have been explored.9–12 Ansa hypoglossus nerve–nerve anastomosis and nerve-muscle implant techniques have also been applied to unilateral vocal fold paralysis.11,13 Despite significant efforts in establishing appropriate reinnervation and function after injury to the recurrent laryngeal nerve, debate continues about the efficacy of these procedures. Reinnervation procedures applied to laryngeal rehabilitation are covered in greater detail in Chapter 68.
The first report of a phonosurgical procedure appeared when Brunings14 introduced the concept of vocal fold medialization by injection of paraffin within the body of the paralyzed fold. This was followed by Payr’s15 description of an external approach for medialization that used a posterior vertical incision through the thyroid lamina, whereby the anterior flap was collapsed inward, resulting in limited medialization. Neither approach gained acceptance.
Almost four decades later, Meurman16 reported a series of patients with vocal cord paralysis in whom external medialization procedures were performed with use of a vertical parasagittal incision in the anterior thyroid cartilage and autologous rib cartilage grafts placed between the thyroid ala and the inner perichondrium. Meurman’s procedure resulted in a high incidence of complications, probably as a result of perichondrial and mucosal perforations occurring with the anterior midline approach.
In the 1960s, Arnold17 reintroduced vocal fold injection but with use of an alloplastic material, polytetrafluoroethylene (Teflon). Over the ensuing years, experience with this material has demonstrated a rising frequency of problems related to granuloma formation.18–20 Subsequently, an absorbable material (absorbable gelatin sponge [Gelfoam]) was applied, allowing temporary vocal fold medialization by injection. Autologous fat has also been used to attempt permanent medialization.21–23 However, the long-term effectiveness of autologous fat has been shown to be unpredictable, with an overall success rate of 62% at 12 months.24 Reports of the use of bovine collagen injections for medialization were initially promising; however, soft tissue response led to variable results with respect to phonatory function.25 At the beginning of this century, micronized AlloDerm Regenerative Tissue Metrix (Cymetra; LifeCell Corporation, Branchburg, NJ) has been applied for vocal fold injection, with early reports indicating improvements in soft tissue response, tissue compliance, and overall phonatory function. Like bovine collagen, Cymetra may last 3 to 9 months and can be used for temporary medialization in patients for whom recovery is likely to occur after recurrent laryngeal nerve injury.26,27
Although numerous modifications of external approaches have been reported,28–32 Isshiki and associates29 were the first to introduce the concept of alloplastic implant material for medialization. Using an external approach with a Silastic implant, these investigators are credited with the ultimate success and popularity of type I medialization thyroplasty. We are proponents of prefabricated implants with sizing systems. Two systems, VoCoM hydroxyapatite implants (Gyrus ACMI, Southborough, MA) and Montgomery Silastic implants (Boston Medical Products Inc, Westborough, MA), are currently available.33 More recently, Gore-Tex strips have been used to maintain vocal fold medialization with and without arytenoid adduction.34–36
Although type I medialization procedures result in dramatic improvement in glottal efficiency and sound production, a small group of patients continue to have difficulty during phonation as a result of a large posterior glottal chink or vocal folds at unequal levels. To address this specific problem, Isshiki and associates29 introduced the arytenoid adduction procedure for unilateral vocal cord paralysis. Placement of a suture around the muscular process of the arytenoid with traction in the direction of the lateral cricoarytenoid and thyroarytenoid muscles results in medial rotation of the arytenoid and downward displacement of the vocal process. The posterior gap is reduced, and the paralyzed vocal folds are placed at equal levels. Arytenoid adduction and related procedures are covered in greater detail in Chapter 67.
The advantages of an external approach to modify vocal fold tension and position without altering the structural components (mucosal fold and underlying muscle body) have expanded the role of laryngeal framework surgery. Isshiki and associates37,38 and Koufman39 have reported their experiences with medialization and tensioning procedures for the management of vocal fold bowing and dysphonia resulting from sulcus vocalis and soft tissue deficits.
Management of Glottal Insufficiency Associated with Unilateral Vocal Fold Motion Impairment and Soft Tissue Deficits
Several procedures are available to manage glottal insufficiency, including vocal fold injection for medialization, medialization thyroplasty, arytenoid adduction, adduction arytenoidopexy,40 and a variety of reinnervation procedures. Selection of the appropriate procedure depends on the duration of symptoms, severity of impairment, presence of anatomic or surgical defect, and potential for recovery. In addition, the patient’s overall condition and life expectancy should be considered before one embarks on a procedure that may predispose the patient to added morbidity.
Severity of impairment can be determined from subjective criteria based on the patient’s symptoms, such as breathiness, aspiration, and exertional intolerance, or on more objective criteria obtained through a variety of tests. Currently available studies for the objective assessment of laryngeal function include perceptual assessment, simple phonatory function tasks such as mean or maximum phonation time, acoustic parameters (i.e., spectrographic analysis, measurement of fundamental frequency, perturbation of frequency and amplitude, signal-to-noise ratio), and measurement of phonatory airflow. The objective assessment of laryngeal function is covered in greater detail in Chapters 57 to 59.
Videostroboscopy remains the most useful subjective and objective test for preoperative and postoperative evaluation of patients with unilateral vocal fold impairment, providing visual assessment of glottal closure and of the status of the mucosal wave. Reinnervation or denervation may be determined from the presence of abnormal or asymmetric mucosal wave patterns during stroboscopy; however, the absence of a mucosal wave does not necessarily imply denervation. Because the mucosal wave is a passive phenomenon established by adequate vocal fold tension and subglottal pressure, the presence of functional motor units is not a prerequisite for the development of the mucosal wave.41
Electromyography (EMG) is the only test available at this time for evaluating the integrity of the laryngeal motor unit in the presence of vocal fold motion impairment. Laryngeal EMG is useful for determining the presence of denervation or reinnervation potentials and is assuming a more active role in the prognostication, timing of intervention, and choice of surgical procedures for the paralyzed larynx.42,43 Despite normal voluntary electric activity, however, vocal fold immobility may be present as the result of laryngeal synkinesis, joint ankylosis, or cicatricial web formation. Distinction among these processes can be made only by palpation of the vocal process during direct laryngoscopy.
With documentation of denervation by EMG, medialization thyroplasty should be considered early in the presence of aspiration or severe dysphonia. If there is evidence of recovery on visual examination or EMG, medialization by injection using a resorbable material such as collagen may be considered as a temporizing procedure. Patients with mild to moderate symptoms of dysphonia only may be counseled to allow spontaneous recovery to occur before considering permanent medialization procedures. Although type I thyroplasty is considered reversible, voice quality after the removal of implants has not been studied.
Prognosis, when combined with results of objective studies, is useful for choice of treatment. Vocal fold paralysis with a favorable prognosis occurs after blunt trauma, endotracheal intubation, idiopathic vocal fold paralysis, and paralysis associated with viral pathogens (Ramsay Hunt syndrome). In this setting, severity of aspiration, dysphonia, and EMG findings can be used to determine the choice of procedure and timing of intervention, as outlined previously. The group of patients with paralysis who have poor prognosis for recovery includes those with injury after complete nerve section during surgical resection of tumor, with invasion of cranial nerves by tumor, with paralysis associated with thoracic aneurysm, and with paralysis due to progressive neurologic disorders. Unlike peripheral recurrent laryngeal nerve injuries, high vagal injuries can result not only in loss of abductor/adductor function but also in loss of cricothyroid muscle function and deafferentation of sensory fibers. In this condition, vocal folds are more likely to be lateralized with marked bowing and atrophic changes. Patients with such injuries generally have greater difficulty with dysphonia, dysphagia, and aspiration. When denervation is documented or history portends a poor outcome, early medialization by thyroplasty is warranted.
Percutaneous medialization by injection should be considered in patients with short life expectancy and aspiration or severe dysphonia. In this situation, the time and expense of medialization thyroplasty or reinnervation procedures might not be justified.
Vocal fold medialization by injection remains a standard procedure for laryngeal rehabilitation. In the absence of arytenoid fixation and when adequate residual vocal fold structure remains to allow needle placement for augmentation, medialization of a paralyzed vocal fold by injection can be performed with use of a variety of materials. Autologous fat and Cymetra are the materials most commonly used today. Cymetra has been shown to provide excellent phonatory results lasting 6 to 12 months with little or no inflammatory response.44 The hyaluronic acid formulations, such as Hylan B gel (Hylaform, Genzyme Biosurgery Inc., Cambridge, MA), used as an injectable material for medialization have also been shown to have favorable viscoelastic properties.45,46 Collagen, although not specifically approved by the U.S. Food and Drug Administration for laryngeal injections, has been shown to be effective for management of vocal fold paralysis, sulcus vocalis, and soft tissue deficits.47 Inflammatory changes associated with soft tissue response after bovine collagen injection can result in greater tissue stiffness and less than satisfactory results. Permanent adverse effects are unlikely to occur, however, because the collagen is ultimately resorbed. The use of polytetrafluoroethylene paste (polytef) for vocal fold injection is discouraged unless long-term patient survival is not anticipated. Adverse effects from development of polytetrafluoroethylene granuloma are well documented and include dysphonia caused by soft tissue reaction and airway obstruction from mass effect.8
Calcium hydroxyapatite (Radiesse dermal filler, Bioform Medical, Inc., San Mateo, CA) has also been introduced as a viable material for injection laryngoplasty with longer duration of effect.48 However, this material has been associated in case reports with a foreign body granulomatous reaction.49 Affected patients have presented with dysphonia, pain, and vocal cord erythema visualized on fiberoptic examination. In our experience and the other reportable case, patients required suspension microlaryngoscopy, which demonstrated multiple irregular mass lesions on the true vocal folds. In both cases, microflaps were made to excise the injectable material. Histologically, a chronic inflammatory infiltrate with surrounding foreign body giant cells was observed. Therefore, we advocate caution in the use of calcium hydroxyapatite as an agent for vocal fold medialization.
Transoral and percutaneous approaches have added a new dimension to the management of vocal fold paralysis.50,51 In most patients, medialization can be accomplished quickly and effectively in the office setting. These procedures are relatively simple and yield immediate results with little discomfort to the patient. If recovery of vocal fold function is likely, fat and polytetrafluoroethylene are usually contraindicated, and alternative methods must be considered. Cymetra can be used as a temporizing measure in this setting. Percutaneous injections are performed without sedation, local anesthesia alone being sufficient. Flexible fiberoptic laryngoscopy is required to visualize position and adequacy of injection. Given their advantages and ease of performance, percutaneous injections are becoming the preferred method in many centers. However, when airway management is a potential problem, injection in a controlled setting during direct laryngoscopy should be considered.
We prefer a lateral, percutaneous approach through the thyroid ala at the level of vocal fold determined by palpation of the thyroid notch and inferior border of the thyroid ala anteriorly (Figs. 66-1 and 66-2). The vocal fold lies perpendicular to this line at the midpoint. Alternatively, an anterior approach can be used through the cricothyroid membrane, with the vocal folds approached from below.50,51 As the needle is inserted, it is angled superiorly and laterally under direct visualization with a flexible fiberoptic nasopharyngoscope. Rees and colleagues52