© Springer International Publishing AG 2018
Christian Sittel and Orlando Guntinas-Lichius (eds.)Neurolaryngologyhttps://doi.org/10.1007/978-3-319-61724-4_1212. Framework Surgery
(1)
Otolaryngology, Head and Neck Surgery, University of Cincinnati, 231 Albert Sabin Way, ML # 0528, Cincinnati, OH 45267-0528, USA
(2)
Otolaryngology, Medical College of Georgia at Augusta University, 1120, 15th Street, Augusta, GA 30912, USA
Abstract
Framework surgery is a widely accepted treatment for vocal fold paralysis; however indications have expanded over time to include other forms of glottal insufficiency. Vocal fold paralysis is the clinical entity that frames our understanding and the historical concepts of what is today known as framework surgery. Over many years adjunct procedures, modifications, and improved clinical evaluation have benefited clinical outcomes. Furthermore, vocal patients can be compared in a number of different measures including visual (videostroboscopy), acoustic (noise-to-harmonic ratio, jitter, shimmer, etc.), and quality of life (voice handicap index, etc.). Initially shaped by our gross anatomic knowledge, fine-tuned by the cover-body theory of vocal fold vibration, and finally research-driven changes based on the biomechanical model will allow us to further improve vocal outcomes in glottal insufficiency with respect to framework surgery.
Keywords
LaryngoplastyThyroplastyVocal fold/cord paralysis/paresisArytenoid adductionGlottal insufficiencyVocal fold bowingPresbylarynges12.1 Introduction
In order to produce a “pure” tone, four physical properties of the vocal fold must be met: (1) inertia (or mass), (2) elasticity (or stiffness), (3) resistance, and (4) symmetry [1]. Unilateral vocal fold paralysis results in a change of all four characteristics. Unilateral vocal fold paralysis can be treated in many ways including voice therapy, reinnervation procedures, injection laryngoplasty, and laryngeal framework surgery. The indications for framework surgery have expanded beyond vocal fold paralysis to include laryngeal paresis and vocal fold atrophy. However, it is vocal fold paralysis that frames our understanding and the historical concepts of these procedures.
12.2 Historical
Laryngology was “born” in the mid-1800s with Garcia’s invention of the laryngeal mirror. Shortly thereafter, Legallois and Longet described laryngeal paralysis [2, 3]. At the time, there was a divergence in medical illustration with the only commonality the “famous wavy or even serpentine cords” demonstrating laryngeal paralysis [2]. In 1881, Semon began to publish his research explaining a sequence of paralytic vocal cords later known as Semon’s law: (1) isolated abductor paralysis, (2) midline position as a result of spastic contracture of adductors, (3) cadaveric position in the stage of complete paralysis of all muscles innervated by the recurrent nerve, and (4) compensation of the paralytic condition by the healthy cord crossing over the midline [2]. In 1857, Grossman describes his concept of laryngeal paralysis as the “median position in complete recurrent paralysis while intermediary position is combined paralysis of both laryngeal nerves” [2]. In 1921, the German Otolaryngological Society entrusted a special commission to study laryngeal paralysis, a detailed questionnaire was published, and many began to reject the validity of Semon’s law, while accepting the greater vulnerability of the abductor fibers based on clinical observation [2].
In 1911, Brünings first developed a method to narrow the glottis using a paraffin injection. This was subsequently abandoned due to adverse side effects of paraffin [4, 5]. Payr in 1915 was the first to describe laryngeal framework surgery using an anteriorly based rectangular thyroid cartilage flap for vocal cord medialization [5]. Meurman in 1952 was the next to utilize a cartilage implant designed from costal rib. Later described were variations of thyroid cartilage by Opheim (1955), Sawashima (1968), and Kramer and Som (1972) [2, 5]. Beginning in 1957, Arnold (innovator of Teflon injection laryngoplasty) described the debate regarding laryngeal paralysis, neural innervation, and clinical signs subsequently writing one of the early comprehensive reviews on laryngeal vocal fold paralysis [2].
These surgeons laid the foundation for Isshiki’s landmark article in 1974 for what is now known as laryngeal framework surgery [4]. Isshiki initially described four types of framework surgery called thyroplasty: type 1 lateral compression, type 2 lateral expansion, type 3 shortening, and type 4 lengthening [4]. Most notable and still most commonly used today for laryngeal paralysis is the type 1 thyroplasty, also known as medialization laryngoplasty (ML). Type 2 was occasionally used in patients with hyperfunctional dysphonia such as plica ventricularis. Type 3 thyroplasty was designed as a relaxation or shortening procedure for abnormally high-pitched voices as in “mutational voice disorder or sulcus vocalis” [4]. Finally, there was type 4 described to recreate the stretching or lengthening for high vagal paralysis, typically in combination with type 1 [4].
Later, Isshiki also described the successful use of an alloplastic implant material (silastic) for medialization laryngoplasty [6]. The adjustability and reversibility of ML was popularized in North America by Koufman and Netterville as they demonstrated highly reproducible voice outcomes in which implants could be adjusted and replaced as necessary [7, 8].
However, the shortcomings of medialization laryngoplasty for unilateral vocal fold paralysis were recognized early on by Isshiki [9]. He described the large posterior glottal chink and vertical height difference, which were not corrected for by ML. In 1978, Isshiki details the first surgery to change the position of the arytenoid cartilage in order to improve vocal outcomes beyond that of medialization laryngoplasty [9]. Over the years, this arytenoid adduction (AA) procedure has undergone several modifications including the posterior anchoring suture [10], cricothyroid subluxation [11], and adduction arytenopexy [12].
Today, framework surgery is the workhorse in surgical management for unilateral vocal fold paralysis. Over time the indications have expanded as we have gained further insight into not only the structural anatomy but also the biomechanical properties of the larynx to further improve vocal outcomes.
12.3 Structural Anatomy
Just as there are gender differences in the vocal folds, not surprisingly, there are also gender differences in the laryngeal cartilaginous framework. The thyroid cartilage diameter in males is thicker than females [13]. The angle of the thyroid ala is greater in the female group (i.e., more obtuse), and the dimension of male larynges is larger on average by 11.4 mm [13]. When measuring the anterior commissure, however, both male and females are consistently in the midline just below the midpoint measuring from the thyroid notch to the inferior border of the thyroid cartilage within 1 mm in both superior and inferior directions [13].
The muscular process of the arytenoid cartilage is located halfway between the roots of the superior and inferior cornu of the thyroid lamina [14]. The recurrent laryngeal nerve is always located deep and medial to the cricothyroid joint and lateral and at the same level of the posterior cricoarytenoid muscle, with no other nerves in this vicinity [14].
The posterior cricoarytenoid muscle is the only abductor; it has two separate identifiable medial and lateral bellies [15]. The medial belly courses superolaterally at an angle of 41° before inserting on the medial aspect of muscular process of the arytenoid [15]. The axis of rotation with horizontal pull has a stronger vertical component compared to the lateral belly with a stronger anterior-posterior component. The lateral belly is at a 33° angle inserting on lateral surface of the muscular process [15]. The motion of the arytenoid rotates about a nearly vertical axis, which allows the vocal process to move more medial without much medial displacement of the arytenoid body [15]. Additionally the arytenoid undergoes a forward tilting motion, which changes the vertical height appreciated clinically in vocal paralysis and corrected post-procedure after arytenoid adduction surgery [15]. Posterior glottic closure improves vocalization following arytenoid adduction by increased subglottic airflow resistance [15].
12.4 Neural Anatomy
Pressman and Kelemen demonstrated the recurrent laryngeal nerve supplies unilaterally all intrinsic muscles of the larynx, while the external branch of the superior laryngeal nerve supplies the external cricothyroid muscle. Both recurrent laryngeal nerves innervate the interarytenoid muscles bilaterally. Hofer and Jeschek demonstrated the internal branch as a purely sensory nerve to reach the mucosa of the aryepiglottic folds [2]. This is a simplistic version of neural anatomy; it is widely accepted that there exists a large variability in cross innervation between left, right, and RLN and SLN differences [16–18].
Laryngeal electromyography (EMG) studies have documented that laryngeal muscles become reinnervated after recurrent laryngeal nerve (RLN) section but rarely after vagus nerve injury [19]. EMG of the thyroarytenoid (TA) and posterior cricoarytenoid (PCA) muscles with vagal injury show denervation and no spontaneous respiratory activation. Woodson also demonstrated both histologic and EMG evidence indicating reinnervation of the TA muscle greater than PCA and showing synkinetic activity especially with respiration. This suggests nerve fibers supposed to supply the PCA had instead synkinetic reinnervation to the TA muscle [19]. This is consistent with clinical observation that patients with vagus nerve injury often demonstrate that the paralyzed vocal fold lies more laterally and with less improvement over time compared to a RLN injury [19]. In unilateral paralysis, preferential reinnervation of adductor muscles is favorable because it enhances glottic closure. Whereas in bilateral paralysis, sustained adduction is unfavorable resulting in airway narrowing during inspiration and therefore more respiratory symptoms despite an acceptable voice [19].
12.5 Mechanics of Voice Production
Van den Berg’s “myoelastic-aerodynamic” model in 1959 was based on experiments of excised larynges; the variables described are initial glottal size, tension of the vocal folds, and subglottal pressure [21]. This model was the first to discover a certain minimal flow was required to vibrate the vocal folds and if the lateral force was too small (persistent glottal gap), the larynges did not respond, i.e., did not vibrate and did not create an acoustic sound [21]. Furthermore, glottal gap was the most important factor to decide the voice quality [22]. When the glottal gap is too small or excessively tight, the voice can be abnormal, rough, or strained. When the stiffness of the vocal folds increases, the range for a normal voice narrows, requiring a higher subglottal pressure [22].
When a paralyzed vocal fold is adequately reconstructed, the stroboscopic findings will reflect improvement in efficient translation of aerodynamic force into an acoustic signal [12]. The stroboscopy may even show normal phase symmetry despite the presence of vocal folds with radically different viscoelastic properties [12]. With uncorrected unilateral paralysis, patients intuitively adjust for glottal incompetence with varying hyperfunctional muscular behaviors [12]. Although static reconstruction realigns the vocal fold to gain aerodynamic competence, the residual perturbation in vocal quality is the result of differential viscoelasticity between the vocal folds [12].
The two biomechanical subunits of the vocal fold can be considered as an anterior and posterior component [23]. By medializing and stiffening the anterior membranous vocal fold of patients with unilateral vocal fold paralysis, glottal efficiency improves [23]. The posterior subunit is the anatomic cartilaginous vocal fold; changes made to this component adjust the position and tension of the vocal process. A midline vocal process results in a better valve to resist airflow leakage, allowing better buildup of subglottic pressure, therefore better acoustic power [23].
The force necessary to lateralize the vocal fold is much less than the force needed to lateralize the vocal process. In Noordizj’s canine model, tension remained relatively unchanged at the mid-membranous region of the vocal fold as suture tension increased [23]. However at the vocal process, vocal fold tension increased with suture tension until 100 g [23]. Suture tensions greater than 100 g did not further increase vocal fold tension at the vocal process. This model helps to explain the suboptimal vocal outcomes with both hypo- and hyper-adduction, clinically experienced during arytenoid adduction surgery [23].
12.6 Patient Evaluation
Glottal incompetence is a common disorder causing both phonatory and swallowing impairments. The voice may be described as weak, breathy with a restricted pitch range [24]. Patients can be virtually asymptomatic or can have a variety of symptoms of glottal insufficiency such as liquid dysphagia, cough, shortness of breath, or exercise intolerance. Clearly, dysphonia is the most common reason for seeking treatment but can present as insufficient loudness, vocal fatigue, globus sensation, effortful voice, impaired singing, etc. [25].
The more specific form of glottal insufficiency, vocal fold paralysis, also varies in clinical presentation. Flaccid paralysis results in altered glottic configuration and arytenoid position [9, 10]. Unilateral vocal fold paralysis can present with structural asymmetries of the membranous folds, vocal fold height, length, and or tension. The arytenoid of the flaccid vocal fold may be rotated so that the vocal process is displaced rostrally and laterally [10]. Because of the angle between the cartilaginous and membranous vocal fold, the patient cannot approximate the vocal folds yielding incomplete glottic closure.
Isshiki’s arytenoid adduction was developed specifically for those patients with a “posterior glottal gap” [9]. Successful AA surgery can increase the apparent length of the paralyzed vocal fold by rotating it toward the visual plane [10]. Another characteristic of laryngeal paralysis is an “anterior tipping” of the arytenoid partially obscuring the vocal fold which is often more severe in complete denervation [10]. This likely occurs by loss of support from the PCA muscles and contributes to inferior displacement of the vocal fold [10]. Contraction of the PCA results in rotation on an axis perpendicular to the AA with posterior “rocking.”
Results of EMG data in laryngeal paralysis may also be beneficial to help guide decision-making. This can come in three general categories: (1) no spontaneous activity and normal recruitment (good prognosis), (2) spontaneous activity in the absence of normal recruitment (poor prognosis), and (3) markedly reduced recruitment with no evidence of denervation (equivocal) [5].
The preoperative videostroboscopy demonstrates the position of the paralyzed vocal fold. According to Netterville, if during phonation the mobile vocal processes contact the immobile side, then medialization alone will be successful [7]. If on the other hand, there is a wide posterior glottic gap then the patient will do better with both ML and AA combined [7, 22].
Improved posterior glottic closure and vocal height discrepancy are two of the most common reasons offered why AA may be considered. However, ML and AA + ML both close the glottal gap with significant improvements in voice quality, but those differences are unique to various independent studies, and no definitive significance has been found looking at vocal outcomes [24, 26, 27]. In reality, the decision based on what surgery is performed is often a function of the physician’s preference and experience [26]. Furthermore, the success of laryngoplastic voice surgery is based on several critical factors, first being the proper patient selection and, second, surgical expertise [22]. Isshiki’s policy for dysphonia second to glottal insufficiency was described as “rough tuning by laryngeal framework surgery and fine tuning by voice therapy” which is an acceptable practice even today [22].
12.7 Indications
According to Netterville, beyond voice outcomes, the second major benefit of medialization laryngoplasty is an improvement in dysphagia with a decrease in aspiration [7]. However, dysphagia is often multifactorial, and improved glottic closure often does not result in better swallowing even with improved voice and a more effective cough [28].
Medialization laryngoplasty with or without arytenoid adduction is useful for narrowing the glottis, especially in the case of a paralyzed vocal cord but is also useful in vocal fold bowing, presbylaryngis, and paresis [29]. Koufman and Postma have described bilateral ML for vocal fold bowing/atrophy [30]. Important considerations for bilateral ML include the following: (1) Overcorrection anteriorly can cause harsh, strained voice. (2) The posterior flange must not contact the arytenoid; otherwise, there is risk of restricting arytenoid motion [30]. Furthermore, ML can be used for a variety of complex anatomic soft tissue deficiencies as a result of cancer resection, trauma, atrophy, and sulcus vocalis. A large prospective study by Zeitels demonstrated the expanded indications for ML, also describing Gore-Tex as a versatile implant ideal for phonosurgical reconstruction [31].
Managing patient expectations is always important. In the case of ML, a consistent pattern of vocal improvement occurs where immediate intraoperative vocal improvements quickly deteriorate as perioperative edema ensues. This needs to be explained to the patient and family so that a somewhat rough immediate postoperative result is expected particularly with the overcorrection required with Gore-Tex ML. Edema is dependent on procedural time and collateral trauma to surrounding soft tissue [12].
12.8 Medialization Laryngoplasty
Generally, framework surgery is recommended and performed under local anesthesia with IV sedation [7]. The size of the implant or degree of adduction is mainly determined by intraoperative voice quality [7]. It is not necessary to use transnasal flexible laryngoscopy with video monitoring; however, it is generally advisable and provides excellent visual feedback during the procedure [7].Intermittent visualization of the larynx by surgeon or assistant can be equally beneficial [7]. If continuous monitoring is used, it is advisable for the patient to receive both topical nasal anesthesia and atropine, an anticholinergic and drying agent [7].
The patient’s head is placed in neutral position, and a 4–5 cm horizontal incision is centered at the mid-thyroid ala, extending just across the midline. Skin is incised, subplatysmal flaps are raised, and the median raphe of the strap muscles is divided. Almost the entire hemilarynx on the involved side is exposed from the thyroid notch superiorly to the cricothyroid membrane inferiorly and midline laterally to the posterolateral border of the thyrohyoid muscle [8]. The perichondrium is elevated from the paralyzed side.
12.8.1 Silastic Implant
The next steps in Isshiki’s type 1 thyroplasty using a silastic implant are as follows. The thyroid window is fashioned. Koufman describes window size in proportion to laryngeal size where men are 5–6 mm high and 12–15 mm wide and women are typically 4–5 mm high and 10–12 mm wide [8]. Half the distance of the thyroid cartilage measured at the thyroid notch determines the position of the upper horizontal window incision [8]. The window is placed parallel to the lower border of the thyroid cartilage, paying close attention to the variability of the anterior third and identifying instead the lower anterior edge and the portion behind the muscular process [7]. This line is angled slightly cephalad (2–4 mm) posteriorly [8]. The window should be as inferior as possible, while still leaving a stable inferior strut [7]. The distance from the anterior commissure to the anterior edge of the window is determined by the shape of the cartilage, which often is a function of gender. More commonly, the laryngeal angle is more obtuse in women (5 mm from the anterior commissure) so more anterior compared to the acute angle in men where the window should be positioned more posteriorly (7 mm from the anterior commissure) [7]. Alternatively or in addition, a pilot hole can be created with an 18G catheter. Next, a 27G needle can be used to gently apply lateral to medial pressure at the suspected level of the true vocal fold. Under direct visual laryngoscopic guidance, the surgeon can verify the position of the glottis prior to creation of the window. Typically the cartilage is calcified, and a 2–3 mm cutting burr using an otologic drill is used to create the window (as an alternative to sharp dissection) and subsequently remove the cartilage island or fragments [7]. The inner perichondrium is widely elevated using a short laryngeal or Penfield elevator. Isshiki’s original description kept the inner perichondrium intact for fear of extrusion of the implant [4]. However, Netterville later described microscopically incising the inner perichondrium to improve medialization without increased rates of extrusion [7]. Thyroarytenoid fascia and distal branches of superior laryngeal vessels are left intact throughout this part of the dissection. Even with the violation of inner perichondrium, a capsule forms around the implant [7]. Next, a blunt instrument or depth gauge is used with vocal feedback from the patient and laryngeal video monitoring to establish the anterior, middle, and posterior depths of the window [7].
Many different variations have been described in carving a silastic implant, first described by Isshiki but further modified by Koufman, Netterville, Bielamowicz, and Montgomery [6–8, 24, 32]. The implant is carved such that the point of maximal medialization is at the inferior edge of the window and away from the ventricle and false vocal folds [7]. Again, because of laryngeal gender differences, the point of maximal medialization measured from the anterior aspect of the window is approximately 10 mm in women and 13–14 mm in men [7]. A 5–6 mm posterior extension can be considered [7]. According to Netterville, the implant can be divided into two pieces to allow a more gentle insertion [7]. Bielamowicz describes a posterior locking flange [24]. Silastic implants are also commercially available as Netterville PhonoForm® Silicone Blocks (Medtronic ENT Jacksonville, FL).
In order to address lengthy intraoperative time and posterior glottic incompetency, Montgomery developed his standardized prosthesis for ML surgery [32, 33]. Available since 1993, the Montgomery® Thyroplasty Implant System is available in 6, 7, 8, 9, and 10 mm female implants and 8, 9, 10, 11, and 12 mm male sizing (Boston Medical Products, Inc., Westborough, MA) [33].
12.8.2 Expanded Polytetrafluorethylene (ePTFE) Implant
Alternatively, the use of expanded polytetrafluoroethylene (ePTFE) or Gore-Tex (W.L. Gore and Associates, Flagstaff, AZ) implants has also been described [34, 35]. This was used to simplify the technique itself and allow incremental adjustment of vocal fold position [34].
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
