Human laryngeal reinnervation: long-term experience with the nerve-muscle pedicle technique

The nerve-muscle pedicle technique for reinnervation of bilateral vocal cord paralysis has now been applied in 45 cases over the past four years. EMG studies, volume flow loop, pulmonary function studies, and the ability to extubate the patient with good to excellent exercise tolerance for day-to-day activity have been used as parameters to judge success or failure of this procedure. To date there has been a primary success rate of 91.1% (41/45) and a long-term success rate of 88.8% (40/45). There has been one serious complication in the entire series and no significant morbidity other than this. Age of the patient and length of paralysis are not necessarily contraindications to the procedure. These results suggest that the nerve-muscle pedicle technique may be successfully used for reinnervation of bilaterally paralyzed vocal cords without sacrificing residual voice, as is the case with arytenoidectomy type procedures.     

Human laryngeal reinnervation

Bilateral vocal cord paralysis is fortunately an uncommon occurrence although the incidence of this injury secondary to external neck trauma is increasing. In general, a patient with this lesion is faced with the choice between adequate airway at the expense of a breathy, weak voice or a fairly good voice with the need for a permanent tracheotomy. In an effort to provide a better solution to this dilemma, an attempt has been made to develop a means of reinnervation of at least one vocal cord without the problems inherent in the usual nerve anastomosis techniques. After extensive preliminary work in dogs the nerve-muscle pedicle technique for reinnervation has been developed. In this procedure the branch of the ansa hypoglossi to the anterior belly of the omohyoid is mobilized and a small block of muscle containing the terminal branches is freed from the muscle proper. Using an approach similar to the Woodman arytenoidectomy, the posterior cricoarytenoid muscle is exposed, and its fibers are partially incised. The previously prepared nerve-muscle pedicle is sutured to it. In the first five patients subjected to this procedure, return of spontaneous abduction of the reinnervated vocal cord was noted between six and eight weeks post surgery. In no case was the voice weakened nor was there any problem with aspiration. All five patients have achieved sufficient air-way so that exercise tolerance for daily activities is adequate without a tracheotomy. The physiologic and histologic background of this technique is discussed in detail.     

Muscle-nerve pedicle laryngeal reinnervation

A muscle-nerve pedicle implantation to the lateral cricoarytenoid muscle as described by Tucker is an alternative to Teflon injection for treating dysphonia due to vocal cord palsy. Improvement in voice was noted in 19 of the 20 (95%) selected patients who were treated by muscle-nerve pedicle reinnervation. These patients have been followed for 6 months to 10 years. Changes in the voice were documented through assessment of high-quality audio tape recordings by three speech pathologists, results of a patient questionnaire, and evaluation by the surgeon. The improvement in voice quality was attributed to reestablishment and maintenance of vocal cord tone and mass, without the vocal cord stiffness usually associated with Teflon injection. The improvements in quality of voice with pedicle implantation are natural and lasting. This technique is proposed as an alternative to Teflon injection in selected cases.     

Surgical anatomy of the recurrent laryngeal nerve: implications for laryngeal reinnervation

Functional laryngeal reinnervation depends upon the precise reinnervation of the laryngeal abductor and adductor muscle groups. While simple end-to-end anastomosis of the recurrent laryngeal nerve (RLN) main trunk results in synkinesis, functional reinnervation can be achieved by selective anastomosis of the abductor and adductor RLN divisions. Few previous studies have examined the intralaryngeal anatomy of the RLN to ascertain the characteristics that may lend themselves to laryngeal reinnervation. Ten human larynges without known laryngeal disorders were obtained from human cadavers for RLN microdissection. The bilateral intralaryngeal RLN branching patterns were determined, and the diameters and lengths of the abductor and adductor divisions were measured. The mean diameters of the abductor and adductor divisions were 0.8 and 0.7 mm, while their mean lengths were 5.7 and 6.1 mm, respectively. The abductor division usually consisted of one branch to the posterior cricoarytenoid muscle; however, in cases in which multiple branches were seen, at least one dominant branch could usually be identified. We conclude that the abductor and adductor divisions of the human RLN can be readily identified by an extralaryngeal approach. Several key landmarks aid in the identification of the branches to individual muscles. These data also indicate the feasibility of selective laryngeal reinnervation in patients who might be candidates for laryngeal transplantation after total laryngectomy.     

Studies of early laryngeal reinnervation

In 1973, Hengerer and Tucker reported a method for laryngeal reinnervation. The purpose of this study is to objectively verify the reinnervation mechanism in 20 dogs with nerve-muscle pedicle grafts. Three months after the experimental operation direct stimulation of the nerve to the neuromuscular pedicle produced an evoked action potential of the posterior cricoarytenoid muscle. This was recorded. Histologic examination of the posterior cricoarytenoid muscle and the nerve-muscle pedicle were performed. The electromyogram revealed in ten of fifteen dogs an excitable action potential. This can be considered as objective evidence for reinnervation from the neuromuscular pedicle.     

Esthesioneuroblastoma: the Northwestern University experience

OBJECTIVE: To review our experience with esthesioneuroblastoma, a rare malignancy of the head and neck. STUDY DESIGN: Retrospective review of Tumor Registry data. METHODS: We performed a computerized search of the Northwestern Memorial Hospital Tumor Registry database from 1981 to 2000. RESULTS: Sixteen patients with esthesioneuroblastoma were identified and analyzed. Their mean age was 42 years. Eleven of 16 patients (69%) had Kadish stage C; 8 patients (50%) had brain involvement at presentation. Craniofacial resection was performed in 13 patients (81%). Fourteen patients received either preoperative or postoperative therapy; radiation therapy was employed in 11 cases and chemotherapy in 4. The actuarial 5-year survival was 60%, and the actuarial 5-year disease-free survival was 33%, with a median follow-up of 4.3 years. Recurrences occurred at a median time of 11 months after diagnosis (2.5 mo-18 y). The first site of failure was locoregional alone in 10 of 12 patients who progressed, and in 6 patients involved the brain or the meninges. Two patients were successfully salvaged. Patients with high-grade tumors had a trend toward work survival. CONCLUSIONS: Esthesioneuroblastoma is a rare tumor that is potentially curable by surgical resection and radiation therapy. However, the rate of local failure is high, and late recurrences are not uncommon. The role of chemotherapy warrants further investigation.     

Motion-specific laryngeal reinnervation using muscle-nerve-muscle neurotization

There is no current treatment method that can reliably restore physiologic movement to a paralyzed vocal fold. The purposes of this study were to test the hypotheses that 1) muscle-nerve-muscle (M-N-M) neurotization can be induced in feline laryngeal muscles and 2) M-N-M neurotization can restore movement to a paralyzed vocal fold. Muscle-nerve-muscle neurotization can be defined as the reinnervation of a denervated muscle via axons that are induced to sprout from nerves within an innervated muscle and that then traverse a nerve graft interposed between it and the target denervated muscle. A paralyzed laryngeal muscle could be reinnervated by axons from its contralateral paired muscle, thus achieving motion-specific reinnervation. Eighteen adult cats were divided into sham, hemilaryngeal-denervated, and M-N-M-reinnervated thyroarytenoid muscle groups. Five of the 6 reinnervated animals had histologic evidence of axons in the nerve graft, 4 of the 6 had evoked electromyographic evidence of crossed reinnervation, and 1 of the 6 had a return of appropriately phased adduction. This technique has great potential and should be further investigated.     

Intrinsic laryngeal muscle reinnervation using the muscle-nerve-muscle technique

OBJECTIVES: This study was performed to investigate the muscle-nerve-muscle reinnervation technique in the larynx, in which a nerve conduit implanted into an innervated muscle conducts axonal sprouting into a denervated muscle while maintaining function of the donor muscle. METHODS: In this study, the muscle-nerve-muscle technique was used to direct superior laryngeal nerve axons to reinnervate intrinsic laryngeal muscles by implanting the recurrent laryngeal nerve stump into the cricothyroid muscle in 8 dogs. In 4 of the dogs, the recurrent laryngeal nerve trunk to the adductor muscles was divided so that all axonal sprouting was directed to the posterior cricoarytenoid muscle. Six-month electromyography data were obtained from 6 of the 8 dogs. RESULTS: All 6 dogs showed evidence of successful reinnervation of the thyroarytenoid or posterior cricoarytenoid muscles with action potentials that corresponded to spontaneous respiratory efforts, while the donor cricothyroid muscles retained their phasic contraction. These responses were obliterated when the recurrent laryngeal nerve conduit was divided. Histologic examination of the intrinsic laryngeal muscles demonstrated successful reinnervation. CONCLUSIONS: The results confirm that intrinsic laryngeal muscles may be successfully reinnervated by the superior laryngeal nerve with the muscle-nerve-muscle technique, without sacrifice of function of the cricothyroid muscle. This method offers an alternative source of appropriately firing axons for laryngeal reinnervation procedures.     

Spontaneous laryngeal reinnervation after recurrent laryngeal or vagus nerve injury

OBJECTIVES: The status of innervation in patients with laryngeal paralysis is somewhat controversial. Electromyographic activity has been frequently documented in the laryngeal muscles of patients with laryngeal paralysis, and animal experiments report a strong propensity for reinnervation after laryngeal nerve injury. However, a study of intraoperative electromyography performed in patients during reinnervation surgery failed to document activity with stimulation of the recurrent laryngeal nerve (RLN). Noting the long-observed differences in the symptoms of patients with vagus nerve injury and those with RLN injury, I hypothesized that reinnervation is influenced by the site of nerve injury. METHODS: Cats were sacrificed at various intervals after resection of 1 cm of either the RLN or the vagus nerve, without any attempt to repair the nerve. RESULTS: Four months after RLN resection, distal nerve biopsy revealed unmyelinated axons scattered through fibrous tissue. By 6 months, myelinated axons were organized, and electromyographic and histologic examination showed preferential reinnervation of the thyroarytenoid muscle. After vagotomy, the RLN was fibrotic and no axons were present. Both the thyroarytenoid and posterior cricoarytenoid muscles were fibrotic and had no electromyographic activity. CONCLUSIONS: The results confirm the strong propensity for laryngeal reinnervation after RLN injury, but not after vagus nerve injury. Preferential reinnervation of adductor muscles may account for a medial position of the paralyzed vocal fold.     

Electromyographic findings in recurrent laryngeal nerve reinnervation

Abductor, adductor, and combined reinnervation procedures have been explored with variable success rates. We describe the experience of a tertiary care center with adductor reinnervation procedures, including preoperative and postoperative videostroboscopy and electromyography (EMG) findings. A retrospective chart review was performed from 1997 to 2001 that included 9 patients. Preoperative and postoperative voice comparison was performed by 3 blinded speech pathologists. Clinical comparisons of videostroboscopy findings for vocal fold bulk, tone, position, presence of gap, and movement are elucidated. The preoperative and postoperative EMG findings are described. In all patients, preoperative EMG revealed a dense, complete denervation of the affected recurrent laryngeal nerve. No movement was noted on videostroboscopy with persistent glottic gap. Reinnervation involved a nerve-muscle pedicle or a direct neurorrhaphy of the ansa cervicalis to the recurrent laryngeal nerve. Voice improvement was noted between 60 days and 3 months after reinnervation. Four postoperative EMG studies were performed. An early postoperative EMG study at 5 months revealed activation of the lateral cricoarytenoid muscle and thyroarytenoid muscle with head-lift. Videostroboscopy showed excellent near-midline static positioning of the vocal fold. Late EMG studies, performed 12 to 16 months after reinnervation, revealed "learning" of these muscles, with new activation on "eee" phonation. We conclude that recurrent laryngeal nerve reinnervation procedures belong in the armamentarium of the laryngologist for the treatment of vocal fold paralysis. The EMG findings reported in this study suggest that ongoing reinnervation allows for activation with phonation in matured neuronal anastomoses. Overall, this procedure results in excellent patient acceptance and near-normal vocal quality.     

Hypoglossal nerve transfer for laryngeal reinnervation: a preliminary study

The hypoglossal nerve is a logical donor nerve for hemilaryngeal reinnervation because 1) its activity coincides with normal laryngeal adduction during speech and deglutition; 2) it is a large nerve with many axons; and 3) donor site morbidity is low. This method of laryngeal reinnervation has not been previously reported. Previous studies using the ansa cervicalis for reinnervation have failed to show spontaneous activity. Hypoglossal-to-recurrent laryngeal nerve anastomosis was performed on a series of 5 dogs. The vocal folds were viewed monthly, with the animals awake, by infraglottic examination through a permanent tracheostomy. One dog failed due to technical error. The remaining 4 dogs began to exhibit spontaneous vocal fold adduction within 2 to 4 months. Vocal fold motion was synchronous with spontaneous tongue motion. Complete glottic closure was seen during swallowing at 3 to 5 months. Intraglottic pressure measurements following reinnervation were normal. Hypoglossal nerve transfer appears to be capable of providing functional adduction to the paralyzed hemilarynx. The potential advantages and disadvantages of this new technique are discussed.     

Vagal stimulation and laryngeal electromyography for recurrent laryngeal reinnervation in children

Ansa-to-recurrent laryngeal nerve (ANSA-RLN) reinnervation procedures are now often first-line treatments for some children with unilateral vocal fold immobility. Although many describe that children with prolonged denervation and true vocal fold atrophy should not undergo this procedure, there has been no gold-standard means of identifying true denervation. Here, we describe a novel technique using evoked vagal electromyography to predict degree of chronic nerve injury prior to recurrent laryngeal nerve reinnervation in children. This is a simple, readily available technique that may play an important role in predicting likelihood of success with pediatric ANSA-RLN reinnervation. Laryngoscope, 130:747–751, 2020     

Long-term results of nerve-muscle pedicle reinnervation for laryngeal paralysis

Between 1976 and 1986, 214 patients with bilateral vocal fold paralysis and 73 patients with unilateral vocal fold paralysis were managed by the author using the nerve-muscle pedicle technique for reinnervation. Follow-up of at least 2 years has been obtained on 202 of the bilaterally and 70 of the unilaterally involved patients. Long-term success has been achieved in 74% of the bilateral group and 88% of the unilateral group. Successful reinnervation of unilateral paralyses usually maintains voice correction indefinitely, but there is late (2 to 5 years postsurgery) deterioration of successful airway restoration in approximately 17% of bilateral cases, which appears to be due to development of cricoarytenoid arthritis.     

Useful landmarks in arytenoid adduction and laryngeal reinnervation surgery

OBJECTIVE: Knowledge of the location of the muscular process of the arytenoid cartilage and the recurrent laryngeal nerve is essential to performing a successful arytenoid adduction and laryngeal reinnervation surgery. We describe external landmarks useful in locating these structures. STUDY DESIGN: Cadaveric laryngeal dissection. METHODS: Posterior laryngeal dissection was performed in 16 human larynges. The position of the muscular process of the arytenoid was measured bilaterally relative to the inferior and superior borders of the thyroid lamina. The recurrent laryngeal nerve was followed distally from slightly below the level of the cricothyroid joint to its genu where its vertical course changes to an oblique intralaryngeal course. RESULTS: The muscular process of the arytenoid was usually found halfway between the roots of the superior and inferior cornu of the thyroid lamina. The recurrent laryngeal nerve was found just deep to the cricothyroid joint and lateral to the posterior cricoarytenoid muscle. There were no other nerves in this area. CONCLUSIONS: This study finds that the superior and inferior borders of the thyroid lamina are useful intraoperative landmarks to locate the muscular process of the arytenoid. The cricothyroid joint provides a good starting point to locate the recurrent laryngeal nerve, which can be identified slightly deeper between it and the posterior cricoarytenoid muscle.     

The effect of patient age on the success of laryngeal reinnervation

The objective of the study was to investigate the influence of patient age on the efficacy of laryngeal reinnervation with ansa cervicalis in unilateral vocal fold paralysis (UVFP) patients. We retrospectively reviewed 349 consecutive UVFP cases of laryngeal reinnervation with ansa cervicalis to the recurrent laryngeal nerve anastomosis. Preoperative and postoperative videostroboscopy, perceptual evaluation, acoustic analysis, maximum phonation time (MPT) and laryngeal electromyography (EMG) data were collected. Gender, age, preoperative EMG status [preoperative voluntary motor unit recruitment (VMUR)] and denervation duration were analyzed in previous multivariable logistic regression analysis. Stratification analysis was performed on patient age in the present study. All patients were divided into four groups according to their age: Group A included patients with an age less than 30 years; Group B, 30–44 years; Group C, 45–59 years; Group D, ≥60 years. Stratification analysis on patient age showed significant differences between Group A and D, Group B and D, Group C and D (P < 0.05), but no significant difference between Group A and B, Group A and C, Group B and C (P > 0.05), respectively, with regard to parameters including glottal closure, overall grade, shimmer, noise-to-harmonics ratio; but there are no significant differences among the four groups with regard to jitter. However, for MPT and postoperative VMUR, there are significant differences among the four groups expect between Group A and B. In addition, glottal closure, perceptual and acoustic parameters, MPT values and VMUR data, were significantly improved postoperatively in each age group (P < 0.01). The data from this study indicate that patient age is an influential factor of the surgical outcome of laryngeal reinnervation for UVFP patients. Laryngeal reinnervation is less effective when patient age is more than 60 years.     

单侧喉返神经损伤神经修复术式探讨

目的　探讨 5种神经修复术治疗单侧喉返神经损伤声带麻痹的疗效。方法　 1993年 1月～ 2 0 0 1年 4月治疗外伤性单侧喉返神经损伤声带麻痹 38例 ,病程从损伤即刻至 2年不等。资料完整者 35例 ,其中行神经减压术 8例、颈襻主支喉返神经吻合术 16例、喉返神经端端吻合术 6例、颈襻神经肌蒂埋植术 3例、颈襻神经植入术 2例。手术前后喉镜、嗓音声学参数、肌电图检查等评价手术效果。结果　病程 4个月内神经减压 5例恢复了正常的声带内收及外展功能 ,4个月以内 1例、以上2例及颈襻主支吻合组、喉返神经端端吻合组则未恢复声带运动。但上述 3种术式均能使喉内收肌获有效的再神经支配 ,满意地恢复声带的肌张力、肌体积、声带振动对称性及正常黏膜波 ,声门闭合良好 ,嗓音恢复正常。颈襻神经肌蒂埋植术及颈襻神经植入术均能改善声嘶 ,但无恢复正常病例。结论　①单侧喉返神经损伤神经修复治疗以神经减压效果最佳 ;②颈襻主支吻合术、喉返神经端端吻合术也能有效地恢复喉的发音功能 ;③喉神经修复术式选择应根据病程、神经损伤程度、类型而定     

Voice quality following laryngeal reinnervation by ansa hypoglossi transfer

Recurrent laryngeal nerve injury resulting in chronic unilateral vocal fold paralysis has been treated traditionally by implantation of various materials into the paralyzed vocal fold. Although the usage of these techniques, especially Teflon®-glycerin paste injection, has been clinically established, they do not restore full functionality to the larynx (abduction, adduction, and vibratory synchronization of the vocal folds). Restoration of these functions, necessary for improved phonation, has been achieved at least on an experimental basis by reinnervation techniques previously described. This study demonstrates excellent human voice quality following reinnervation of the vocal folds in two cases using ansn hypoglossi-recurrent laryngeal nerve anastomosis. Although the reinnervated vocal fold neither abducted nor adducted, it presented itself in the midline for precise apposition with the nonparalyzed cord. Voice data were analyzed within a single subject experimental design at the following intervals: preoperatively, immediately postoperatively, midterm, and long-term (3 and 6 years). The data was analyzed by subjective and objective means, including acoustics and electroglottography. Patient selection, surgical techniques, results, and implications are reviewed.