Surgical correction of arytenoid position is an effective means of managing patients with posterior glottic incompetence due to laryngeal paralysis [1-3]. Many surgeons avoid arytenoid procedures because they perceive this approach to be difficult and fraught with complications . Additionally, most patients with laryngeal paralysis respond to either injection augmentation or open thyroplasty; however, persisting posterior gap is a common reason for failure of medialization procedures. A review of 20 cases of failed medialization laryngoplasty found that 12 patients had persisting posterior gaps, which were subsequently corrected by arytenoid adduction . Other clinical reports, as well as experimental studies in animals, indicate that arytenoid adduction is much more effective than type-I thyroplasty in closing the posterior glottis [6, 7]. Successful management of patients with flaccid laryngeal paralysis requires attention to the posterior gap; therefore, it is important that all surgeons who care for patients with laryngeal paralysis be familiar with the indications for arytenoid repositioning. With knowledge of local anatomy and understanding of key technical principles, the procedure can be mastered, with a complication rate comparable to that of medialization thyroplasty , A study of type-I thyroplasty in 237 patients compared outcomes with and without arytenoid adduction. Average operative time and hospital stay were both longer with arytenoid adduction (73 vs 45 minutes and 1.8 vs 1.1 days), but there was no significant difference in incidence or severity of complications . In fact, neither of the two patients who required emergency tracheotomy for airway obstruction had undergone arytenoid adduction.
The symptoms of laryngeal paralysis vary greatly, ranging from no symptoms to glottal incompetence, with aphonia and aspiration. When the arytenoid cartilage of the paralyzed vocal fold is in an internally rotated position, the vocal process is near the midline, so that during phonatory closure the membranous vocal folds are essentially parallel and separated by a gap ofless than 3 mm. Many such patients are able to compensate by increasing adduction of the normal vocal fold. Those patients who do not develop effective compensation can usually achieve a good voice after either vocal fold augmentation or medialization thyroplasty, however, when the vocal process of the arytenoid is not near the midline, the membranous folds diverge posteriorly, so that there is a wide glottic gap during phonation and sometimes aspiration during swallowing. Such a gap is too wide for closure to be achieved by injection augmentation or thyroplasty. The vertical position of the paralyzed vocal fold may also affect function. For example, the vocal process may be displaced superiorly and laterally, due to external rotation of the cricoarytenoid joint. In other patients, the arytenoid "sags" forward, so that the vocal process is at or below the phonatory plane. In either case, the vocal folds do not make contact, even when they seem to be adjacent when viewed from above. Surgical correction of arytenoid position can be an effective tool for correcting the position of the vocal process in both the horizontal and vertical planes.
When viewed from above, by mirror examination or endoscopy, the vocal folds appear to open and close in an axial plane, pivoting at the anterior commissure, like opposing windshield wipers; however, vocal fold motion is considerably more complex. The chief major moving parts of the larynx are the arytenoid cartilages, which articulate with the cricoid cartilage in shallow, ball, and socket-type joints and rotate around variable axes. The membranous vocal folds are rigidly fixed anteriorly to the thyroid cartilage and posteriorly to the vocal processed of the arytenoid cartilages. Opening and closing of the glottis is achieved by motion of the arytenoids, so that the vocal processes move medially and laterally. The membranous vocal folds are passively dragged to and from, varying the angle between the vocal folds. The only muscles that abduct the vocal fold are the posterior cricoarytenoid muscles (PCA), which originate on the posterior cricoid and insert on the muscular process of the arytenoid. During PCA contraction, the arytenoids rotate externally, around a nearly vertical axis, so that the vocal process moves up and out (Fig. 10.1) . All other muscles that insert on the arytenoid exert a closing force, but with differing vectors of pull. For example, contraction of the lateral cricoarytenoid muscle rotates the arytenoid internally so that the vocal processes move inwardly and caudally. The "criss-crossing" fibers of the unpaired interarytenoid muscle internally rotate the arytenoid in a coronal plane, pulling the apices medially. Because each muscle has a different vector of force, the cricoarytenoid joint has a wide range of motion. The vocal process moves superolaterally with active abduction and inferomedially with active adduction; however, a paralyzed vocal fold need not lie within the trajectory of normal motion. Many authors have reported superior displacement of the paralyzed vocal fold [11,12]. But in some patients with flaccid paralysis, the vocal process is displaced caudally as the arytenoid tips forward, due to lack of support from the PCA muscle . Cadaver specimens from patients with laryngeal paralysis have demonstrated caudal displacement of the paralyzed vocal fold and shift of the conus elasticus to a horizontal plane [14, 15]. During phonation, the mobile vocal fold can compensate to some degree for vertical misalignment. But a vertical gap that cannot be easily appreciated during routine laryngeal exam can substantially impair phonation.
What Is a Posterior- Gap?
The term "posterior gap" is somewhat misleading. A true posterior gap normally occurs in phonation. The vocal processes are apposed, but there is usually a small gap between the arytenoid bodies [16,17]. Frequently, this space is not visible during transoral examination, as it is obscured by overhanging posterior glottic tissue, and may actually be filled in by soft tissue. The posterior glottis can be best observed in patients with a tracheotomy, by retrograde flexible endoscopy via the stoma. This posterior gap is usually aerodynamically insignificant, as induction of glottic vibration requires parallel apposition of the membranous folds .
In patients with posterior glottic incompetence, the maximal glottic opening is actually between the vocal processes. In the abducted glottis, the posterior glottic walls actually converge posterior to the vocal process. When a vocal fold is paralyzed in this abducted position, the posterior body of its arytenoid blocks the normal one from achieving vocal process contact; thus, the vibratory portion of the vocal folds are not parallel, but diverge posteriorly. Another factor that contributes to glottic incompetence in flaccid paralysis is that the paralyzed vocal process is not stable but can be driven laterally by the force of the mobile vocal fold during adduction. Finally, as mentioned above, the vocal process may be vertically displaced, above or below the plane of phonation.
The first surgical approaches to the arytenoid were to remove or lateralize it in patients with airway obstruction due to bilateral laryngeal paralysis . In 1948, Morrison described surgical displacement of the arytenoid to medialize a unilaterally paralyzed vocal fold. He called his operation the "reverse King" procedure . Via a lateral approach to the larynx, the cricoarytenoid joint was opened, and the capsule was completely divided to mobilize the arytenoid. The arytenoid was then moved medially and fixed to the cricoid cartilage with a suture. In 1977 Ballenger stated that the Morrison pro cedure had "been supplanted by intrachordal injections" . Indeed, Teflon injection, as described by Arnold in 1962, was the prevailing treatment for unilateral laryngeal paralysis for many years . Over time, it became apparent that Teflon injection had a significant incidence of delayed complications, specifically, granuloma formation. Additionally, injection laryngoplasty was ineffective in patients with a large glottal gap and did not close the posterior glottis. In 1975 Isshiki et al. reported the used of thyroplasty type-I as an alternative to injection . Their procedure was similar in concept to the thyroid cartilage flap that had been described by Payr in 1915 . Within a few years, type-I thyroplasty had become widely accepted as the procedure of choice for unilateral laryngeal paralysis; however, it was noted that type-I thyroplasty, like injection laryngoplasty, was not effective in closing the posterior glottis.
In 1978 Isshiki reported arytenoid adduction as a means of treating patients with large glottal gaps . They noted that the normal motion of the arytenoid was rotational, rather than translational, and that rotating the arytenoid through its natural range of motion was much more effective in closing the glottis than attempting to move the entire arytenoid medially along the cricoid (Fig. 10.2). This proce-
dure gained acceptance more slowly than type-I thyroplasty, due to concerns about technical difficulty and complications. In addition, the procedure, as originally described, included dividing the cricothyroid joint and opening the cricoarytenoid joint. In a series of 12 arytenoid adduction procedures, Slavit and Maragos found over-rotation of the arytenoid and shortening of the vocal fold in 1 patient, presumably due to destabilization of the arytenoid by opening the joints . Netterville et al. also noted unfavorable voice results that seemed to result from over-rotation or anterior displacement of the larynx . Woodson subsequently reported modification of the technique to preserve the cricothyroid and cricoarytenoid joints, which prevented anterior arytenoid dislocation and resulted in more consistent vocal process position, and to dissect lateral to the strap muscles, to improve exposure and reduce risk to the carotid . In addition to improved vocal quality, mean phonation time was substantially increased in these patients. Subsequently, larger series have demonstrated the efficacy of arytenoid adduction [8,26].
Zeitels et al.  has developed a different approach to arytenoid surgery, the "adduction arytenoidopexy," which is similar in concept to the "reverse King procedure" developed by Morrison . In adduction arytenopexy, the lateral wall of the cricoarytenoid joint is opened widely. A single suture is then placed through posterior cricoid and through the body or the muscular process of the arytenoid. The suture is positioned so that it draws arytenoid posteriorly, superiorly, and medially. The goal is to mimic combined motion of lateral cricoarytenoid, thyroarytenoid, and interarytenoid muscles. Published results in a series of 12 patients supports the efficacy of the procedure . Zeitels et al. developed his approach to overcome perceived shortcomings of arytenoid adduction. Firstly, they states that the normal vocal fold contour is straight during adduction, and arytenoid adduction "hyperrotates" the arytenoid, so that the vocal process is too medial, resulting in a persisting chink persists behind the vocal processes. Adduction arytenopexy was designed to translate the entire arytenoid medially, rather than rotate it. The goal of his proce dure is to medialize and lengthen the paralyzed vocal fold, and to achieve a straight vocal fold edge; however, there is considerable evidence that during normal phonation the vocal fold is not straight and there is usually a problem behind the vocal processes [16,17].
Adduction arytenopexy is technically more difficult and less standardized than arytenoid adduction. The suture should pass through either the body or vocal process of the arytenoid. There are no guidelines for determining exact suture placement, as the description of the procedure states, "The completion of the stabilizing suture can be done in several ways and is individualized to each patient and to surgeon preference." Furthermore, the vector of pull is essentially localized in the posterior glottis, in the coronal plane with narrowly spaced points of fixation, and lateral stability is sacrificed by the opening of the joint. The sutures exert a medial force on the body of the arytenoid but do not prevent slight rotation with lateral displacement of the vocal process during phona-tory closure. At present, arytenoid adduction is more widely used than adduction arytenopexy.
Medialization procedures, which focus on displacing the anterior vocal fold, do have some influence on the position of the vocal process, since the two structures are connected; however, the torque achieved by medial displacement of the muscular portion of the vocal fold is inadequate to effect significant internal rotation of the arytenoid cartilage. The vector of force in medialization is linear, toward the midline, while arytenoid motion is actually rotational. The rationale for arytenoid repositioning surgery is to restore the phonatory position of the paralyzed vocal process and to stabilize the arytenoid cartilage in internal rotation, preventing passive displacement during glottal closure. The objective is effective glottal closure, with a minimum of compensatory hyperfunction. Repositioning of the arytenoid does not restore bulk to the atrophic muscles of the paralyzed vocal fold, and so concomitant medialization procedure may be required.
The mechanics of the arytenoid adduction procedure have been studied in cadaver larynges. Three-dimensional coordinates of laryngeal cartilages were determined by CT at rest and after arytenoid adduction, and then rotation and translation was computed in three planes, using rigid body mechanics . The arytenoid rotates about a nearly vertical axis with the vocal process moving downward and medially. The paralyzed vocal fold generally appears shorter than the mobile side . In patients, as viewed from above, arytenoid adduction appears to lengthen the paralyzed vocal fold; however, research in cadaver larynges indicates that the arytenoid adduction procedure does not actually lengthen the vocal fold, but moves the vocal process caudally . It is this vertical component of motion that is endo-scopically perceived as a length change. An abducted vocal fold appears shorter when viewed from above, because the vocal process has moved rostrally; the vocal fold slopes upward, out of the plane of the image. In arytenoid adduction the vocal process is displaced caudally, toward the level of the anterior commissure, so that the vocal fold is parallel to the image plane, and apparently longer. With bilateral arytenoid adduction in cadaver larynges, vocal processes are pulled together, and resistance to air flow is increased .
Other studies in cadaver larynges explored the influence of varying the site of attachment of the anterior sutures, and have also assessed induced phonation. Results indicate that fine adjustment of the arytenoid in three planes can be achieved by varying the anterior site of attachment of the adduction suture. Such adjustments are more evident in phonatory function than in visually detectable variation in position [30,31].
Zeitels et al.  has compared the results of arytenoid adduction and adduction aryteno-pexy in cadaver larynges, by performing the two procedures on opposing sides of several specimens. The vertical positions of the two vocal folds were compared by placing a ruler between the vocal processes. The length of each vocal fold was measured with calipers and the angle of the arytenoid measured from a cranial perspective. Finally, the displacement of the arytenoid along the cricoid was measured with calipers. The findings indicated no vocal fold length change with arytenoid adduction, consistent with prior studies, but a 2-mm increase after adduction arytenopexy. Vertical position measurement indicated that the vocal process was 2 mm higher on the side of the adduction arytenopexy . The downward displacement of the vocal process is appropriate when the paralyzed vocal fold lies above the plane of glottic closure. But in some patients with flaccid laryngeal paralysis, the arytenoid "sags" anteriorly. In such cases the arytenoid adduction cannot correct, and may exacerbate, the vertical gap between the vocal processes during closure. A second suture can be attached to the vocal process and secured posteriorly, to provide posterior suspension, essentially replacing the force of normal resting tone in the PCA muscle .
Arytenoid adduction is usually performed under local anesthesia. This allows the vocal results to be assessed intraoperatively, and avoids the problem of an endotracheal tube that blocks adduction of the glottis; however, it can be performed under general anesthesia by placing sufficient tension on the adducting suture to achieve maximal anterior motion of the muscular process. This generally results in appropriate adduction of the vocal fold when the patient is awake and extubated. A general anesthetic is most useful in anxious or immature patients or in those where surgical dissection is hindered by scarring, obesity, or presence of a tracheotomy.
A horizontal incision is made in the neck over the lower thyroid ala, similar to the incision used for thyroplasty, but extending somewhat further posteriorly. In most cases a thyroplasty window should be created at this point. Such a window is a very good route for passing the adduction suture. It also prepares the field for a thyroplasty, either planned, or to correct a persistent anterior gap after arytenoid adduction.
The most effective exposure of the posterior thyroid lamina is achieved by dissecting lateral to the cervical strap muscles. This allows the entire laryngotracheal complex to be rotated as a unit, providing broad access to the posterior larynx. This approach also decreases the risk to the carotid sheath, since the larynx is displaced away from that structure (Fig. 10.3). The omohyoid muscle can be grasped to aid in rotating the larynx during dissection, but stable rotation during arytenoid surgery is provided by sturdy hook on the superior cornu or the hyoid bone. If necessary, the thyrohyoid ligament can be divided to allow more rotation and greater exposure.
To access the cricoarytenoid joint, it is necessary to go through the inferior pharyngeal constrictor muscle and then displace the mucosa of the pyriform sinus. One can accomplish this by sharply dividing the inferior constrictor muscle along the posterior edge of the thyroid lamina. The nerve to the cricothyroid can usually be located and preserved, although the functional consequences of sacrificing this nerve would be minimal. Another means of managing the constrictors is to incise the perichondrium anterior to their attachment, and reflecting that flap posteriorly . The pyriform sinus is then dissected off the PCA muscle and displaced superiorly and anteriorly, until the muscular process of the PCA is located. It can usually be easily identified by following the
fibers of the PCA muscle until they converge at the muscular process (Fig. 10.4). If exposure is unsatisfactory, it can be improved by removing a portion of the posterior edge of the thyroid cartilage. An alternate surgical approach is the posterior thyroplasty window, as described by Maragos .
Patients with paralysis due to a vagus nerve injury frequently have dysphagia with aspiration, due to unilateral pharyngeal constrictor paralysis in the face of continued activity in the bilaterally innervated cricopharyngeus muscle. In such cases, a cricopharyngeal myotomy can dramatically improve swallowing function, and is easily added to this procedure, by an inferior extension of the incision of the through the inferior constrictor muscle, onto the cricoid cartilage .
As originally described, the next steps in the procedure would be to disarticulate the cricothyroid joint and then open the cricoarytenoid joint; however, most authors now recommend preserving natural biomechanics by leaving these structures intact. The muscular process is exposed by further refection of the pyriform sinus. It is important not to violate that mucosa by tearing during dissection or puncturing it with a suture needle, as a leak could contaminate the field and possibly result in a fistula; thus, identification and preservation of the margins of the pyriform sinus can be the most challenging part of the procedure. An awake
and cooperative patient can be instructed to produce positive intraoral pressure (Say "puppy"), which distends and clearly delineates the pyriform.
The adducting suture should be placed horizontally through the muscular process, assuring that a good "bite" of cartilage is obtained. The suture should be tied securely, leaving two long ends.
A key step in the procedure is to clear a tunnel as a path for the suture to follow to the anterior larynx. If the suture passes through any connective tissue or muscle en route, that tissue could act as a pulley that alters the vector of force and prevents adequate adduction. If the suture is through muscle or tendon, the resulting torque will be inadequate. It is important that this dissection be through the paraglottic space, medial to perichondrium and external to muscle, to decrease the chance ofhemorrhage.
Once the tunnel is created, a small hemostat can be used to pass one end of the arytenoid suture through the tunnel and out through the thyroplasty window. Ultimately, both suture segments will be passed forward and secured to the lower thyroid cartilage. But initially only one segment is passed forward, so that alternating tension on the two sutures can be used to test the efficacy of rotation. The anterior tension should be sufficient to move and maintain the vocal process as far anteriorly as possible. This generally requires very little force. Increasing tension after the arytenoid has reached maximal internal rotation is counterproductive, and can distort the vocal fold, or even result in avulsion of the suture. The larynx should be observed from above by endoscopy to confirm adequate vocal process adduction. If satisfactory position is not achieved, the suture insertion should be re-evaluated and may need to be revised. If the adduction is satisfactory, then the other end of the suture is passed forward through the tunnel. Adequate position is reassured, and then secured with a surgeon's knot, and several throws. This author has found that the most satisfactory means of securing the suture anteriorly is to bring one end through the thyroplasty window and the other through the cricothyroid space, below the posterior inferior corner of the window. This permits subsequent medialization of the anterior vocal fold with an implant (Fig. 10.5); however, research by Inagi and colleagues in cadaver larynges suggests that optimal medialization is achieved by anchoring the suture as far inferiorly and anteriorly as possible in the thyroid cartilage .
A posterior suspension suture should be considered if the arytenoid is tipped anteriorly, with caudal displacement of the vocal process . The location of this suture on the arytenoid should be superior to that of the primary suture. Careful examination and palpation of the arytenoid will reveal a ridge that runs from the muscular process to the apex of the arytenoid. The lateral cricoarytenoid muscle attaches to the anterior surface of this ridge, while the interarytenoid muscle attaches posteriorly. The suspension suture should be attached to this ridge, about 1 cm away from the primary suture. Inferior traction should be applied to rock the arytenoid posteriorly, while observing the larynx endoscopically to assure improved laryngeal posture, without lateral displacement of the vocal process. This suture is then secured inferiorly by passing one end around or through the inferior cornu of the thyroid cartilage.
A drain is prudent, as well as overnight observation, since bleeding into the paraglottic space can cause airway obstruction. This author has encountered one patient in 15 years
who required emergency tracheotomy because of an obstructing hematoma. Another potential source obstruction is herniation of pyriform mucosa, as reported by Weinman and Mara-gos . They noted this complication after arytenoid adduction via a posterior thyroplasty window, and have found that tacking down the pyriform mucosa is effective in preventing airway obstruction.
Surgical adjustment of the arytenoid cartilage is a valuable tool in the surgical armamentarium for the treatment of laryngeal paralysis. While this approach is technically more difficult and requires more operative time than a type-I thyroplasty, it is much more effective in patients with a large glottic gap and posterior glottic incompetence.
1. Isshiki N, Tanabe M, Masaki S (1978) Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol HeadNeck Surg 14:555-558
2. Woodson GE (1997) Cricopharyngeal myotomy and arytenoid adduction in the management of combined laryngeal and pharyngeal aralysis. Otolaryngol HeadNeck Surg 117:339-343
3. Zeiteis SM, Hochman I, Hillman RE (1998) Adduction arytenoidopexy: a new procedure for paralytic dysphonia with implications for implant medialization. Ann Otol Rhinol Laryngol (Suppl) 107:173
4. Koufman JA, Isaacson G (1991) Laryngoplastic phonosurgery. Otolaryngol Clin North Am:ii5i-
Was this article helpful?
Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!