Original Research
Revision Transcervical Medialization Laryngoplasty for Unilateral Vocal Fold Paralysis
Otolaryngology– Head and Neck Surgery 1–6 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599815585091 http://otojournal.org
Noah P. Parker, MD1, Anca M. Barbu, MD1, Robert E. Hillman, PhD1, Steven M. Zeitels, MD1, and James A. Burns, MD1
No sponsorships or competing interests have been disclosed for this article.
Keywords vocal cord, paralysis, dysphonia, revision, laryngoplasty
Abstract Objective. To identify patterns of failure following transcervical medialization laryngoplasty for unilateral vocal fold paralysis and describe indications and revision techniques for optimal vocal outcomes. Study Design. Case series with chart review. Setting. Tertiary care center. Subjects and Methods. Thirty-nine consecutive patients between January 2005 and April 2014 undergoing transcervical revision of failed primary medialization laryngoplasty were identified. Demographics, etiology, stroboscopic assessment, and surgical techniques were recorded. Patient self-assessment using the Voice-Related Quality-of-Life (VRQOL) questionnaire and objective acoustic and aerodynamic assessments performed pre- and postoperatively were analyzed using t tests for paired comparisons. Results. Thirty-nine patients underwent 48 transcervical revision surgeries. Median follow-up was 14.6 months from time of final revision surgery. Indications included anterior glottic incompetence (38/48, 79%), posterior glottic incompetence (20/48, 42%), glottic overclosure (8/48, 17%), and/or decreased phonatory pliability (12/48, 25%). A combination of findings was present in 21 (44%) surgeries. Revision techniques included either anterior augmentation, arytenopexy, and cricothyroid subluxation (alone or in combination) in 46 of 48 (96%) patients or partial implant removal alone in 2 patients. Seven patients (18%) required multiple revisions. A complete set of voice parameters was available for 22 patients, and statistically significant improvements included VRQOL scores, fundamental frequency in females, jitter, noise-to-harmonic ratio, and mean airflow rate. Conclusion. Patterns of failure in patients with suboptimal phonatory function after transcervical medialization laryngoplasty included persistent glottic incompetence, glottic overclosure, and decreased vocal fold pliability. Revision transcervical medialization surgery, guided by individualized consideration of vocal fold position and surface pliability, can improve phonatory outcomes.
Received February 19, 2015; revised March 30, 2015; accepted April 13, 2015.
T
ranscervical medialization laryngoplasty (ML) has remained a primary surgical option for patients with symptomatic unilateral vocal fold paralysis (UVFP) since its introduction by Payr in 19151 and its further development and popularization by Isshiki et al2 in the 1970s. While a variety of implant materials and techniques have been introduced, the surgical goal continues to be repositioning the immobile vocal fold to restore glottal competence. Medializing the anterior, membranous glottis often involves augmenting the paraglottic space with various alloplastic materials,2-6 while repositioning the posterior glottis involves moving either the vocal process7 or the entire arytenoid.8 Patients who fail to have adequate improvement of phonatory function following primary ML can be challenging to evaluate and treat. The interplay between voice perception, overall glottal competence, stroboscopic findings, and patients’ vocal needs obligates the surgeon to consider all of these factors to optimize results from revision surgery. The purpose of this study is to identify patterns of failure following primary ML for UVFP and to describe indications and targeted revision techniques for optimal vocal outcomes.
Materials and Methods After institutional review board approval through Massachusetts General Hospital, a database was queried for all patients treated 1 Department of Surgery, Harvard Medical School, and Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, Boston, Massachusetts, USA
Corresponding Author: James A. Burns, MD, Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, One Bowdoin Square, 11th Floor, Boston, MA 02114, USA. Email: [email protected]
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
2
Otolaryngology–Head and Neck Surgery
with ‘‘medialization laryngoplasty,’’‘‘arytenoid adduction,’’ and/or ‘‘cricothyroid subluxation’’ between January 2005 and April 2014 for UVFP. Patients were included if they underwent transcervical revision of primary ML and if complete medical documentation of initial consultation, stroboscopic evaluation, operative details, and follow-up was available for review. Patients were excluded if any of the above data were incomplete or if there was another confounding laryngeal or neurologic process affecting phonatory function. Patients with incomplete objective voice measures were included in revision results, but only patients with a complete set of preand postrevision objective voice measures were considered in voice outcomes reporting. Data extraction included demographics, etiology of paralysis, time elapsed since primary surgery, reason for failure after primary ML, revision technique, complications, and voice outcomes. When considering the reasons for failure, factors involving patients’ history, subjective auditory assessment, and stroboscopy findings were considered. Stroboscopy was performed using the KayPentax Digital Video Stroboscopy System and either a 70° transoral rigid laryngoscope or a flexible transnasal video nasopharyngolaryngoscope (KayPentax, Montvale, New Jersey). Imaging was confirmed in each idiopathic case, and computed tomography was routinely obtained to further evaluate patients for whom primary procedures were performed at outside hospitals. The surgical techniques for primary ML using GoreTex,2,6,9 as well as that for arytenopexy and cricothyroid subluxation, have been previously described.8,10 Planning revision surgery involved careful consideration of the patients’ persistent symptoms and vocal demands, breathing limitations and requirements for physical activity, and the identification of sites of glottal incompetence, glottal overclosure, and/or phonatory mucosal pliability. Voice outcomes were reported for patients who had completed pre- and postoperative voice evaluations using the Voice-Related Quality-of-Life (VRQOL) questionnaire11 and objective acoustic and aerodynamic analyses. Acoustic and aerodynamic testing was done in a sound-treated room using the KayPentax Computerized Speech Laboratory and Phonatory Aerodynamic System with associated software modules. Acoustic measures included jitter, shimmer, noiseto-harmonic ratio (NHR), and fundamental frequency (f0). Aerodynamic measures included airflow rate and air pressure. Vocal function outcomes were assessed by comparing pre- and postoperative measures using t tests for paired comparisons with the level of significance at P .05.
Results In total, 256 MLs were performed during the 9-year study period, of which 48 were revision surgeries performed for 39 patients who met criteria for data extraction. All patients undergoing revision ML satisfied the inclusion criteria without need for exclusion. Table 1 lists demographic and etiological data for the patient cohort. Median follow-up was 14.6 months from the final revision ML (range, 1 day to 75
Table 1. Demographics and Etiology of Unilateral Vocal Fold Paralysis in Patients Undergoing Revision Transcervical Medialization Laryngoplasty (n = 39).a Characteristic Age at surgery, mean (SD), y Sex Male Female Side of paralysis Left Right Etiology Iatrogenic (n = 30) Thyroidectomy Skull base surgery Esophagectomy Aortic arch repair Anterior approach to cervical spine Intracranial surgery Pneumonectomy Recurrent laryngeal nerve section Thymectomy Carotid endarterectomy Mediastinal malignancy excision Idiopathic (n = 6) Cervical metastasis infiltration (n = 2) Intubation-related (n = 1)
Value 59.3 (14.5) 19 20 25 14
8 4 3 3 3 2 2 2 1 1 1
a
Values are presented as numbers unless otherwise indicated.
months). Follow-up for some patients required distant travel so that only immediate follow-up was available. Stroboscopy identified 4 structural findings that warranted consideration during revision: anterior glottic incompetence (38/48, 79%), posterior glottic incompetence (20/48, 42%), glottic overclosure (8/48, 17%), and/or decreased phonatory mucosal pliability (12/48, 25%). A combination of findings was present for 21 (44%) of the revision surgeries, which is why the additive total percent exceeds 100%. This revision ML cohort of 48 surgeries includes 23 (48%) following the authors’ primary surgeries and 25 (52%) following primary surgeries performed by outside surgeons. An injection medialization laryngoplasty was performed prior to revision ML in 14 of 48 (29%) surgeries. Accounting for the 208 patients treated by the authors with primary ML during the study period (256 total MLs minus the 48 revision MLs) and the 23 revision surgeries performed following primary surgery by the authors (25 revisions were following primary surgery by outside surgeons), the in-house revision rate was 23 of 208 (11%). Anterior augmentation, arytenopexy, and cricothyroid subluxation (alone or in combination) were the predominant revision ML techniques (46/48, 96%; see Figure 1). Anterior augmentation included implanting the ipsilateral or contralateral paraglottic space with Gore-Tex, complete removal of the
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
Parker et al
3
Figure 1. Venn diagram showing revision transcervical medialization laryngoplasty procedures involving anterior augmentation, arytenopexy, and/or cricothyroid subluxation. *Anterior augmentation includes procedures placing Gore-Tex implant in the ipsilateral and/ or contralateral paraglottic space, complete explantation with reimplantation of Gore-Tex implant, or adjustment of the position of previously placed Gore-Tex implant.
originial implant with reimplantation of Gore-Tex, or adjusting previously placed Gore-Tex. The remaining 2 procedures required partial removal of an oversized implant. Ten patients (26%) underwent postrevision ML injection laryngoplasty with hyaluronic acid gel (Restylane; Medicis Aesthetics, Scottsdale, Arizona), and 4 of these patients went on to receive an additional transcervical revision ML. A total of 7 patients (18%) required multiple revisions (5 patients required 2 revisions; 2 patients required 3 revisions). There were 4 complications (4/48, 8%): 2 hematomas in patients undergoing arytenopexy (1 required reoperation), 1 patient developed postoperative cervical cellulitis treated with antibiotics, and 1 patient developed postoperative subcutaneous emphysema that resolved with conservative management. All complications were following open revision ML; there were no complications following injection laryngoplasty. Eighty-five percent (33/39) of patients had subjective improvement in voice at final follow-up. Acoustic, aerodynamic, and VRQOL data were available for analysis in 22 patients, and the results are shown in Table 2. Statistically significant improvements included total score, physical domain score, and social-emotional domain score from the VRQOL data; f0 in females, jitter, and NHR from the acoustic data; and mean airflow rate from the aerodynamic data. All other changes were not significant, including f0 in males.
Discussion Optimizing voice results during revision ML can be challenging. Few reports in the literature focus on revision ML,12-18
and all reports are retrospective in nature with small cohorts of patients. Our revision rate of 11% is consistent with previously reported revision rates of 4.5% to 16%.13,14,16,18,19 A detailed and thorough examination, including stroboscopy with or without imaging, is critical to choosing the appropriate technique for revision ML that matches the site of failure. In a study of 20 patients, Woo et al14 found that a persistent posterior glottal gap was the most common reason for failure (55%), followed by improper implant size or position causing anterior incompetence. In a study of 16 patients, Cohen et al16 found that the most common reasons for closure problems were undercorrection (75%), persistent posterior glottal gap (17%), and anterior overclosure (8%). Our larger cohort of 39 patients had one or a combination of the following reasons for failure in descending order of prevalence: anterior glottal incompetence, posterior glottal incompetence, glottic overclosure, and/or phonatory mucosal stiffness. This is the first study to include recognition of decreased phonatory mucosal pliability as an important factor in determining voice outcomes in revision ML. A treatment algorithm is presented in Figure 2 that includes consideration of decreased phonatory mucosal pliability. While the algorithm in Figure 2 compartmentalizes various findings and couples those findings with surgical options, many patients have multiple factors prompting consideration of combined techniques and staged procedures. To illustrate the point, we present a 49-year-old woman with dysphonia following total thyroidectomy who failed to vocally improve after a calcium hydroxylapatite injection at an outside institution. Stroboscopy showed right vocal fold immobility, a fulllength glottic gap, and a subepithelial space collection of material, presumed to be superficial injectate, limiting mucosal wave propagation (see Figure 3a). A combination of arytenopexy, Gore-Tex implant placement, and cricothyroid subluxation was planned to improve glottic closure. The patient was counseled that a staged procedure would likely be necessary to address the dual problem of incomplete glottic closure and loss of pliability. After undergoing arytenopexy and cricothyroid subluxation to improve glottic closure, the patient underwent transoral endoscopic removal of the subepithelial material (Figure 3b). While voice was improved, there was persistent dysphonia and a small mid-cord gap, so another transcervical revision ML was performed with additional paraglottic GoreTex placement. At final follow-up, the patient reported nearnormal voice, and stroboscopy showed complete glottic closure (Figure 3c). Her VRQOL scores improved from 40 at initial consultation to 83 following final revision. This case highlights the complex nature of revision ML, which requires an understanding of anatomic aspects of anterior and posterior glottic closure, as well as the physiologic requirements of vocal fold pliability to ultimately achieve a satisfactory vocal result. Revision ML is further complicated by scarring from previous dissection within the paraglottic space for implant placement. While limited fibrosis around a Gore-Tex implant has been shown20 compared with silastic implants21 in animal
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
4
Otolaryngology–Head and Neck Surgery
Table 2. Acoustic, Aerodynamic, and Quality-of-Life Outcome Data for Patients Undergoing Revision Medialization Laryngoplasty (n = 22). Measure VRQOL Total score Physical Social-emotional Acoustic F0 (males), Hz F0 (females), Hz Jitter Shimmer NHR Aerodynamic Air flow rate, L/s Air pressure, CmH2O
Presurgery, Mean (SD)
Postsurgery, Mean (SD)
Mean Change
P Value
43.6 (20.2) 40.4 (19.7) 48.2 (24.9)
65.6 (25.3) 63.9 (25.4) 68.0 (29.2)
122.0 123.5 119.8
.001a .000a .010a
161 (48) 218 (42) 5.245 (2.6) 6.546 (3.8) 0.357 (0.19)
132 (15) 189 (31) 2.973 (1.6) 5.988 (2.5) 0.260 (0.20)
–30 –29 –2.272 –0.558 –0.097
.133 .040a .001a .370 .031a
0.481 (0.37) 11.471 (5.3)
0.322 (0.28) 12.984 (5.9)
–0.159 11.513
.005a .259
Abbreviations: NHR, noise-to-harmonic ratio; VRQOL, Voice-Related Quality of Life. a Significant change at P .05.
Figure 2. Algorithm depicting surgical decision making for revision transcervical medialization laryngoplasty. *Augmentation can also be performed contralaterally or bilaterally based on patient-specific anatomy. AA, arytenoid adduction; CT, cricothyroid.
models, intraoperative adjustments can be difficult during revision surgery. Such tissue changes may explain why patients undergoing revision ML may ultimately require further revisions. In a study of 96 patients, Anderson et al15 noted that 22.9% of patients who underwent revision ML required additional procedures to reach a satisfactory voice result, including 12 injection medializations and 14 additional open revision MLs either anteriorly or posteriorly. In our series, one-third of patients required additional procedures after revision ML,
including 10 patients receiving injection medialization (4 of whom went on to receive transcervical re-revision ML) and 3 patients who required transcervical re-revision ML without initial injections. Patients undergoing revision ML should be counseled about the potential higher risk for multiple surgeries to optimize voice outcomes. Subjective voice outcomes pre- and postrevision ML have previously been reported by 4 studies.13,16-18 Subjective voice improvement was noted in 80% of patients by Maragos13 and
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
Parker et al
5
Figure 3. (a) Initial stroboscopy image showing incomplete glottic closure during phonation and white material within the superficial lamina propria of the right vocal fold. (b) Intraoperative image showing removal of stiff subepithelial material from the right vocal fold. (c) Stroboscopy image after 2 revisions and removal of subepithelial injectate showing complete glottic closure.
in 85% of patients by Lundeberg et al.18 Cohen et al,16 using the glottal closure index22 recorded pre- and postoperatively, reported that 81% of 16 patients significantly improved postoperatively. Andrews et al17 used their own measures of dysphonia severity, impact of dysphonia, and vocal effort pre- and postoperatively and found significant improvements in all measures. The group also reported significant improvement in each parameter of the grade, roughness, breathiness, asthenia, strain (GRBAS) scale23 postrevision.17 This study herein showed subjective voice improvement in 85% of patients at final follow-up, and VRQOL data showed significant improvements. Objective voice outcomes reported in 3 studies pre- and postrevision ML show less consistent improvements.14,17,18 In 7 patients, Woo et al14 reported improvement in airflow approaching statistical significance but no difference in phonation time, fundamental frequency, jitter, shimmer, and NHR. In 6 patients, Andrews et al17 noted significant improvement in maximum phonation time, and in 5 patients, Lundeberg et al18 found significant improvements in maximum phonation duration, pitch range, and fundamental frequency. Anderson et al reported preoperative (not postoperative) aerodynamic measures in an effort to find associations with vocal outcome and identified a lower NHR as predictive of poorer outcome but no correlation with mean phonation time or mean flow rates. Acoustic and aerodynamic data reported in this study showed more consistent improvements and support our patient-reported results. Figure 2 presents an algorithmic approach for revision ML that attempts to match clinical observations with management strategies. As depicted, anterior glottic incompetence and glottic overclosure offer more straightforward management options. In such cases, revision ML can include addition of implant into the ipsilateral or contralateral paraglottic space, complete explantation with reimplantation of implant, or adjustment of the position of a previously placed implant. Vocal fold injection with resorbable gels can be used to improve voice and set appropriate expectations prior to revision ML or can be used in a more permanent fashion before or after an open revision. In this series, 6 of 10 patients receiving postrevision vocal fold injection did not require further open revision, implying lasting benefit. Similar to results reported by Andrews et al,17 this series does not report any implant infections. Posterior glottic gaps have been shown
to predict vocal function outcomes following primary ML,24 and the importance of recognizing posterior incompetence in revision cases is no different. Surgical revision in such cases, assuming good abduction of the contralateral cord and no excessive respiratory requirements, often requires arytenoid repositioning. Implants that are too large can create a convex shape of the medial vocal fold edge that reduces propagation of the mucosal wave during phonation. This occurrence directly affects the ipsilateral vocal fold and can indirectly impede function of the contralateral vocal fold by creating an abnormal glottic closure pattern during phonation. Decreased phonatory mucosal pliability from previously unrecognized pathology, such as phonotrauma, iatrogenic damage, presbyphonia, or sulcus, may adversely affect voice results despite proper position of the medialized immobile vocal fold. Recognizing decreased pliability can prevent additional unnecessary surgery that attempts to further medialize the immobile vocal fold. While treatment options for decreased pliability are limited, it may be possible to augment the paraglottic space (such as with fat implant or repositioning of the existing implant) to realign the medial edge and mobilize any remaining pliable tissue to improve the glottic sound source. The algorithm in Figure 2 includes an option for considering reduced pliability during ML. In the illustrative case mentioned previously, the patient’s vocal fold pliability was improved by removal of a previously injected material that was too superficial. Several weaknesses of this study are readily apparent. The retrospective design of this study presents a weakness that could be overcome by designing a hypothesis-driven prospective study that considered factors contributing to the need for revision medialization laryngoplasty. Complete pre- and postoperative voice measures for all patients within the cohort would potentially make conclusions more powerful. Patients in this study had variable follow-up due the nature of referral patterns from outside institutions and their ability to travel for repeat examinations. Adequate longterm follow-up for all patients would enhance the results.
Conclusions Patterns of failure in patients with suboptimal phonatory function after transcervical medialization laryngoplasty included persistent glottic incompetence, glottic overclosure,
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
6
Otolaryngology–Head and Neck Surgery
and decreased vocal fold pliability. Closure of a persistent posterior glottic gap and proper assessment of phonatory pliability were important factors for revision in this patient cohort. Revision transcervical medialization surgery, guided by individualized consideration of vocal fold position and surface pliability, can improve phonatory outcomes as assessed by subjective (VRQOL) as well as objective acoustic and aerodynamic measures of vocal function. Author Contributions Noah P. Parker, wrote article, collected data, analyzed data, revised article, final approval, accountable for all aspects of the work; Anca M. Barbu, analyzed revision data, revised article, final approval, accountable for all aspects of the work; Robert E. Hillman, collected voice data, analyzed voice data, revised article, final approval, accountable for all aspects of the work; Steven M. Zeitels, analyzed revision data, revised article, final approval, accountable for all aspects of the work; James A. Burns, designed study, analyzed revision data, revised article, final approval, accountable for all aspects of the work.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Payr E. Plastik am Schildknorpel zur Behebung der Folgen Einsetiger Stimmbandlahmung. Dtsch Med Wochensch. 1915; 43:1265-1270. 2. Isshiki N, Morita H, Okamura H, et al. Thyroplasty as a new phonosurgical technique. Acta Otolaryngol (Stockh). 1974;78: 451-457. 3. Koufman JA. Laryngoplasty for vocal cord medialization: an alternative to Teflon. Laryngoscope. 1986;96:726-730. 4. Montgomery WM, Blaugrund SM, Varvares MA. Thyroplasty: a new approach. Acta Otol Rhinol Laryngol. 1993;102:571-579. 5. Cummings CW, Purcell LL, Flint PW. Hydroxylapatite laryngeal implants for medialization: preliminary report. Ann Otol Rhinol Laryngol. 1993;102:843-851. 6. Hoffman HT, McCulloch TM. Anatomic considerations in the surgical treatment of unilateral laryngeal paralysis. Head Neck. 1996;18:174-187. 7. Isshiki N, Tanabe M, Sawada M. Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol. 1978;104:555-558. 8. Zeitels SM, Hochman I, Hillman RE. Adduction arytenopexy: a new procedure for paralytic dysphonia with implications for implant medialization. Ann Otol Rhinol Laryngol. 1998;107:2-24.
9. Zeitels SM. Adduction arytenopexy with medialization laryngoplasty and cricothyroid subluxation: a new approach to paralytic dysphonia. Op Tech Otolaryngol Head Neck Surg. 1999;10:9-16. 10. Zeitels SM, Hillman RE, Desloge RB, et al. Cricothyroid subluxation: a new innovation for enhancing the voice with laryngoplastic phonosurgery. Ann Otol Rhinol Laryngol. 1999;108:1126-1131. 11. Hogikian ND, Sethuraman G. Validation of an instrument to measure voice-related quality of life (VRQOL). J Voice. 1999; 13:557-569. 12. Koufman JA, Postma GN. Revision laryngoplasty. Op Tech Otolaryngol. 1999;10:61-65. 13. Maragos NE. Revision thyroplasty. Ann Otol Rhinol Laryngol. 2001;110:1087-1092. 14. Woo P, Pearl AW, Hsiung M, et al. Failed medialization laryngoplasty: management by revision surgery. Otolaryngol Head Neck Surg. 2001;124:615-621. 15. Anderson TD, Spiegel JR, Sataloff RT. Thyroplasty revisions: frequency and predictive factors. J Voice. 2003;17:442-448. 16. Cohen JT, Bates DD, Postma GN. Revision Gore-Tex medialization laryngoplasty. Otolaryngol Head Neck Surg. 2004;131: 236-240. 17. Andrews BT, Van Daele DJ, Karnell MP, et al. Evaluation of open approach and injection laryngoplasty in revision thyroplasty procedures. Otolaryngol Head Neck Surg. 2008;138:226-232. 18. Lundeberg MR, Flint PW, Purcell LL, et al. Revisions medialization thyroplasty with hydroxylapatite implant. Laryngoscope. 2001;121:999-1002. 19. Netterville JL, Stone RE, Luken ES, et al. Silastic medialization and arytenoid adduction: the Vanderbilt experience. Ann Otol Rhinol Laryngol. 1993;102:413-424. 20. Ustundag E, Boyaci Z, Keskin G, et al. Soft tissue response of the larynx to silicone, Gore-Tex, and irradiated cartilage implants. Laryngoscope. 2005;115:1009-1014. 21. Paniello RC, Dahm JD. Reversibility of medialization laryngoplasty: an experimental study. Ann Otol Rhinol Laryngol. 1997;106:902-908. 22. Duke SG, Salmon J, Blalock PD, et al. Fascia augmentation of the vocal fold: graft yield in the canine and preliminary clinical experience. Laryngoscope. 2001;111:759-764. 23. Hirano M. Clinical examination of the voice. In: Arnold GE, Winckle F, Wyke BD, eds. Disorders of Human Communication. New York, NY: Springer-Verlag; 1981:81-84. 24. Omori K, Slavit DH, Kacker A, et al. Quantitative criteria for predicting thyroplasty type 1 outcomes. Laryngoscope. 1996; 106:689-693.
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
Revision Transcervical Medialization Laryngoplasty for Unilateral Vocal Fold Paralysis
Otolaryngology– Head and Neck Surgery 1–6 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599815585091 http://otojournal.org
Noah P. Parker, MD1, Anca M. Barbu, MD1, Robert E. Hillman, PhD1, Steven M. Zeitels, MD1, and James A. Burns, MD1
No sponsorships or competing interests have been disclosed for this article.
Keywords vocal cord, paralysis, dysphonia, revision, laryngoplasty
Abstract Objective. To identify patterns of failure following transcervical medialization laryngoplasty for unilateral vocal fold paralysis and describe indications and revision techniques for optimal vocal outcomes. Study Design. Case series with chart review. Setting. Tertiary care center. Subjects and Methods. Thirty-nine consecutive patients between January 2005 and April 2014 undergoing transcervical revision of failed primary medialization laryngoplasty were identified. Demographics, etiology, stroboscopic assessment, and surgical techniques were recorded. Patient self-assessment using the Voice-Related Quality-of-Life (VRQOL) questionnaire and objective acoustic and aerodynamic assessments performed pre- and postoperatively were analyzed using t tests for paired comparisons. Results. Thirty-nine patients underwent 48 transcervical revision surgeries. Median follow-up was 14.6 months from time of final revision surgery. Indications included anterior glottic incompetence (38/48, 79%), posterior glottic incompetence (20/48, 42%), glottic overclosure (8/48, 17%), and/or decreased phonatory pliability (12/48, 25%). A combination of findings was present in 21 (44%) surgeries. Revision techniques included either anterior augmentation, arytenopexy, and cricothyroid subluxation (alone or in combination) in 46 of 48 (96%) patients or partial implant removal alone in 2 patients. Seven patients (18%) required multiple revisions. A complete set of voice parameters was available for 22 patients, and statistically significant improvements included VRQOL scores, fundamental frequency in females, jitter, noise-to-harmonic ratio, and mean airflow rate. Conclusion. Patterns of failure in patients with suboptimal phonatory function after transcervical medialization laryngoplasty included persistent glottic incompetence, glottic overclosure, and decreased vocal fold pliability. Revision transcervical medialization surgery, guided by individualized consideration of vocal fold position and surface pliability, can improve phonatory outcomes.
Received February 19, 2015; revised March 30, 2015; accepted April 13, 2015.
T
ranscervical medialization laryngoplasty (ML) has remained a primary surgical option for patients with symptomatic unilateral vocal fold paralysis (UVFP) since its introduction by Payr in 19151 and its further development and popularization by Isshiki et al2 in the 1970s. While a variety of implant materials and techniques have been introduced, the surgical goal continues to be repositioning the immobile vocal fold to restore glottal competence. Medializing the anterior, membranous glottis often involves augmenting the paraglottic space with various alloplastic materials,2-6 while repositioning the posterior glottis involves moving either the vocal process7 or the entire arytenoid.8 Patients who fail to have adequate improvement of phonatory function following primary ML can be challenging to evaluate and treat. The interplay between voice perception, overall glottal competence, stroboscopic findings, and patients’ vocal needs obligates the surgeon to consider all of these factors to optimize results from revision surgery. The purpose of this study is to identify patterns of failure following primary ML for UVFP and to describe indications and targeted revision techniques for optimal vocal outcomes.
Materials and Methods After institutional review board approval through Massachusetts General Hospital, a database was queried for all patients treated 1 Department of Surgery, Harvard Medical School, and Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, Boston, Massachusetts, USA
Corresponding Author: James A. Burns, MD, Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, One Bowdoin Square, 11th Floor, Boston, MA 02114, USA. Email: [email protected]
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
2
Otolaryngology–Head and Neck Surgery
with ‘‘medialization laryngoplasty,’’‘‘arytenoid adduction,’’ and/or ‘‘cricothyroid subluxation’’ between January 2005 and April 2014 for UVFP. Patients were included if they underwent transcervical revision of primary ML and if complete medical documentation of initial consultation, stroboscopic evaluation, operative details, and follow-up was available for review. Patients were excluded if any of the above data were incomplete or if there was another confounding laryngeal or neurologic process affecting phonatory function. Patients with incomplete objective voice measures were included in revision results, but only patients with a complete set of preand postrevision objective voice measures were considered in voice outcomes reporting. Data extraction included demographics, etiology of paralysis, time elapsed since primary surgery, reason for failure after primary ML, revision technique, complications, and voice outcomes. When considering the reasons for failure, factors involving patients’ history, subjective auditory assessment, and stroboscopy findings were considered. Stroboscopy was performed using the KayPentax Digital Video Stroboscopy System and either a 70° transoral rigid laryngoscope or a flexible transnasal video nasopharyngolaryngoscope (KayPentax, Montvale, New Jersey). Imaging was confirmed in each idiopathic case, and computed tomography was routinely obtained to further evaluate patients for whom primary procedures were performed at outside hospitals. The surgical techniques for primary ML using GoreTex,2,6,9 as well as that for arytenopexy and cricothyroid subluxation, have been previously described.8,10 Planning revision surgery involved careful consideration of the patients’ persistent symptoms and vocal demands, breathing limitations and requirements for physical activity, and the identification of sites of glottal incompetence, glottal overclosure, and/or phonatory mucosal pliability. Voice outcomes were reported for patients who had completed pre- and postoperative voice evaluations using the Voice-Related Quality-of-Life (VRQOL) questionnaire11 and objective acoustic and aerodynamic analyses. Acoustic and aerodynamic testing was done in a sound-treated room using the KayPentax Computerized Speech Laboratory and Phonatory Aerodynamic System with associated software modules. Acoustic measures included jitter, shimmer, noiseto-harmonic ratio (NHR), and fundamental frequency (f0). Aerodynamic measures included airflow rate and air pressure. Vocal function outcomes were assessed by comparing pre- and postoperative measures using t tests for paired comparisons with the level of significance at P .05.
Results In total, 256 MLs were performed during the 9-year study period, of which 48 were revision surgeries performed for 39 patients who met criteria for data extraction. All patients undergoing revision ML satisfied the inclusion criteria without need for exclusion. Table 1 lists demographic and etiological data for the patient cohort. Median follow-up was 14.6 months from the final revision ML (range, 1 day to 75
Table 1. Demographics and Etiology of Unilateral Vocal Fold Paralysis in Patients Undergoing Revision Transcervical Medialization Laryngoplasty (n = 39).a Characteristic Age at surgery, mean (SD), y Sex Male Female Side of paralysis Left Right Etiology Iatrogenic (n = 30) Thyroidectomy Skull base surgery Esophagectomy Aortic arch repair Anterior approach to cervical spine Intracranial surgery Pneumonectomy Recurrent laryngeal nerve section Thymectomy Carotid endarterectomy Mediastinal malignancy excision Idiopathic (n = 6) Cervical metastasis infiltration (n = 2) Intubation-related (n = 1)
Value 59.3 (14.5) 19 20 25 14
8 4 3 3 3 2 2 2 1 1 1
a
Values are presented as numbers unless otherwise indicated.
months). Follow-up for some patients required distant travel so that only immediate follow-up was available. Stroboscopy identified 4 structural findings that warranted consideration during revision: anterior glottic incompetence (38/48, 79%), posterior glottic incompetence (20/48, 42%), glottic overclosure (8/48, 17%), and/or decreased phonatory mucosal pliability (12/48, 25%). A combination of findings was present for 21 (44%) of the revision surgeries, which is why the additive total percent exceeds 100%. This revision ML cohort of 48 surgeries includes 23 (48%) following the authors’ primary surgeries and 25 (52%) following primary surgeries performed by outside surgeons. An injection medialization laryngoplasty was performed prior to revision ML in 14 of 48 (29%) surgeries. Accounting for the 208 patients treated by the authors with primary ML during the study period (256 total MLs minus the 48 revision MLs) and the 23 revision surgeries performed following primary surgery by the authors (25 revisions were following primary surgery by outside surgeons), the in-house revision rate was 23 of 208 (11%). Anterior augmentation, arytenopexy, and cricothyroid subluxation (alone or in combination) were the predominant revision ML techniques (46/48, 96%; see Figure 1). Anterior augmentation included implanting the ipsilateral or contralateral paraglottic space with Gore-Tex, complete removal of the
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
Parker et al
3
Figure 1. Venn diagram showing revision transcervical medialization laryngoplasty procedures involving anterior augmentation, arytenopexy, and/or cricothyroid subluxation. *Anterior augmentation includes procedures placing Gore-Tex implant in the ipsilateral and/ or contralateral paraglottic space, complete explantation with reimplantation of Gore-Tex implant, or adjustment of the position of previously placed Gore-Tex implant.
originial implant with reimplantation of Gore-Tex, or adjusting previously placed Gore-Tex. The remaining 2 procedures required partial removal of an oversized implant. Ten patients (26%) underwent postrevision ML injection laryngoplasty with hyaluronic acid gel (Restylane; Medicis Aesthetics, Scottsdale, Arizona), and 4 of these patients went on to receive an additional transcervical revision ML. A total of 7 patients (18%) required multiple revisions (5 patients required 2 revisions; 2 patients required 3 revisions). There were 4 complications (4/48, 8%): 2 hematomas in patients undergoing arytenopexy (1 required reoperation), 1 patient developed postoperative cervical cellulitis treated with antibiotics, and 1 patient developed postoperative subcutaneous emphysema that resolved with conservative management. All complications were following open revision ML; there were no complications following injection laryngoplasty. Eighty-five percent (33/39) of patients had subjective improvement in voice at final follow-up. Acoustic, aerodynamic, and VRQOL data were available for analysis in 22 patients, and the results are shown in Table 2. Statistically significant improvements included total score, physical domain score, and social-emotional domain score from the VRQOL data; f0 in females, jitter, and NHR from the acoustic data; and mean airflow rate from the aerodynamic data. All other changes were not significant, including f0 in males.
Discussion Optimizing voice results during revision ML can be challenging. Few reports in the literature focus on revision ML,12-18
and all reports are retrospective in nature with small cohorts of patients. Our revision rate of 11% is consistent with previously reported revision rates of 4.5% to 16%.13,14,16,18,19 A detailed and thorough examination, including stroboscopy with or without imaging, is critical to choosing the appropriate technique for revision ML that matches the site of failure. In a study of 20 patients, Woo et al14 found that a persistent posterior glottal gap was the most common reason for failure (55%), followed by improper implant size or position causing anterior incompetence. In a study of 16 patients, Cohen et al16 found that the most common reasons for closure problems were undercorrection (75%), persistent posterior glottal gap (17%), and anterior overclosure (8%). Our larger cohort of 39 patients had one or a combination of the following reasons for failure in descending order of prevalence: anterior glottal incompetence, posterior glottal incompetence, glottic overclosure, and/or phonatory mucosal stiffness. This is the first study to include recognition of decreased phonatory mucosal pliability as an important factor in determining voice outcomes in revision ML. A treatment algorithm is presented in Figure 2 that includes consideration of decreased phonatory mucosal pliability. While the algorithm in Figure 2 compartmentalizes various findings and couples those findings with surgical options, many patients have multiple factors prompting consideration of combined techniques and staged procedures. To illustrate the point, we present a 49-year-old woman with dysphonia following total thyroidectomy who failed to vocally improve after a calcium hydroxylapatite injection at an outside institution. Stroboscopy showed right vocal fold immobility, a fulllength glottic gap, and a subepithelial space collection of material, presumed to be superficial injectate, limiting mucosal wave propagation (see Figure 3a). A combination of arytenopexy, Gore-Tex implant placement, and cricothyroid subluxation was planned to improve glottic closure. The patient was counseled that a staged procedure would likely be necessary to address the dual problem of incomplete glottic closure and loss of pliability. After undergoing arytenopexy and cricothyroid subluxation to improve glottic closure, the patient underwent transoral endoscopic removal of the subepithelial material (Figure 3b). While voice was improved, there was persistent dysphonia and a small mid-cord gap, so another transcervical revision ML was performed with additional paraglottic GoreTex placement. At final follow-up, the patient reported nearnormal voice, and stroboscopy showed complete glottic closure (Figure 3c). Her VRQOL scores improved from 40 at initial consultation to 83 following final revision. This case highlights the complex nature of revision ML, which requires an understanding of anatomic aspects of anterior and posterior glottic closure, as well as the physiologic requirements of vocal fold pliability to ultimately achieve a satisfactory vocal result. Revision ML is further complicated by scarring from previous dissection within the paraglottic space for implant placement. While limited fibrosis around a Gore-Tex implant has been shown20 compared with silastic implants21 in animal
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
4
Otolaryngology–Head and Neck Surgery
Table 2. Acoustic, Aerodynamic, and Quality-of-Life Outcome Data for Patients Undergoing Revision Medialization Laryngoplasty (n = 22). Measure VRQOL Total score Physical Social-emotional Acoustic F0 (males), Hz F0 (females), Hz Jitter Shimmer NHR Aerodynamic Air flow rate, L/s Air pressure, CmH2O
Presurgery, Mean (SD)
Postsurgery, Mean (SD)
Mean Change
P Value
43.6 (20.2) 40.4 (19.7) 48.2 (24.9)
65.6 (25.3) 63.9 (25.4) 68.0 (29.2)
122.0 123.5 119.8
.001a .000a .010a
161 (48) 218 (42) 5.245 (2.6) 6.546 (3.8) 0.357 (0.19)
132 (15) 189 (31) 2.973 (1.6) 5.988 (2.5) 0.260 (0.20)
–30 –29 –2.272 –0.558 –0.097
.133 .040a .001a .370 .031a
0.481 (0.37) 11.471 (5.3)
0.322 (0.28) 12.984 (5.9)
–0.159 11.513
.005a .259
Abbreviations: NHR, noise-to-harmonic ratio; VRQOL, Voice-Related Quality of Life. a Significant change at P .05.
Figure 2. Algorithm depicting surgical decision making for revision transcervical medialization laryngoplasty. *Augmentation can also be performed contralaterally or bilaterally based on patient-specific anatomy. AA, arytenoid adduction; CT, cricothyroid.
models, intraoperative adjustments can be difficult during revision surgery. Such tissue changes may explain why patients undergoing revision ML may ultimately require further revisions. In a study of 96 patients, Anderson et al15 noted that 22.9% of patients who underwent revision ML required additional procedures to reach a satisfactory voice result, including 12 injection medializations and 14 additional open revision MLs either anteriorly or posteriorly. In our series, one-third of patients required additional procedures after revision ML,
including 10 patients receiving injection medialization (4 of whom went on to receive transcervical re-revision ML) and 3 patients who required transcervical re-revision ML without initial injections. Patients undergoing revision ML should be counseled about the potential higher risk for multiple surgeries to optimize voice outcomes. Subjective voice outcomes pre- and postrevision ML have previously been reported by 4 studies.13,16-18 Subjective voice improvement was noted in 80% of patients by Maragos13 and
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
Parker et al
5
Figure 3. (a) Initial stroboscopy image showing incomplete glottic closure during phonation and white material within the superficial lamina propria of the right vocal fold. (b) Intraoperative image showing removal of stiff subepithelial material from the right vocal fold. (c) Stroboscopy image after 2 revisions and removal of subepithelial injectate showing complete glottic closure.
in 85% of patients by Lundeberg et al.18 Cohen et al,16 using the glottal closure index22 recorded pre- and postoperatively, reported that 81% of 16 patients significantly improved postoperatively. Andrews et al17 used their own measures of dysphonia severity, impact of dysphonia, and vocal effort pre- and postoperatively and found significant improvements in all measures. The group also reported significant improvement in each parameter of the grade, roughness, breathiness, asthenia, strain (GRBAS) scale23 postrevision.17 This study herein showed subjective voice improvement in 85% of patients at final follow-up, and VRQOL data showed significant improvements. Objective voice outcomes reported in 3 studies pre- and postrevision ML show less consistent improvements.14,17,18 In 7 patients, Woo et al14 reported improvement in airflow approaching statistical significance but no difference in phonation time, fundamental frequency, jitter, shimmer, and NHR. In 6 patients, Andrews et al17 noted significant improvement in maximum phonation time, and in 5 patients, Lundeberg et al18 found significant improvements in maximum phonation duration, pitch range, and fundamental frequency. Anderson et al reported preoperative (not postoperative) aerodynamic measures in an effort to find associations with vocal outcome and identified a lower NHR as predictive of poorer outcome but no correlation with mean phonation time or mean flow rates. Acoustic and aerodynamic data reported in this study showed more consistent improvements and support our patient-reported results. Figure 2 presents an algorithmic approach for revision ML that attempts to match clinical observations with management strategies. As depicted, anterior glottic incompetence and glottic overclosure offer more straightforward management options. In such cases, revision ML can include addition of implant into the ipsilateral or contralateral paraglottic space, complete explantation with reimplantation of implant, or adjustment of the position of a previously placed implant. Vocal fold injection with resorbable gels can be used to improve voice and set appropriate expectations prior to revision ML or can be used in a more permanent fashion before or after an open revision. In this series, 6 of 10 patients receiving postrevision vocal fold injection did not require further open revision, implying lasting benefit. Similar to results reported by Andrews et al,17 this series does not report any implant infections. Posterior glottic gaps have been shown
to predict vocal function outcomes following primary ML,24 and the importance of recognizing posterior incompetence in revision cases is no different. Surgical revision in such cases, assuming good abduction of the contralateral cord and no excessive respiratory requirements, often requires arytenoid repositioning. Implants that are too large can create a convex shape of the medial vocal fold edge that reduces propagation of the mucosal wave during phonation. This occurrence directly affects the ipsilateral vocal fold and can indirectly impede function of the contralateral vocal fold by creating an abnormal glottic closure pattern during phonation. Decreased phonatory mucosal pliability from previously unrecognized pathology, such as phonotrauma, iatrogenic damage, presbyphonia, or sulcus, may adversely affect voice results despite proper position of the medialized immobile vocal fold. Recognizing decreased pliability can prevent additional unnecessary surgery that attempts to further medialize the immobile vocal fold. While treatment options for decreased pliability are limited, it may be possible to augment the paraglottic space (such as with fat implant or repositioning of the existing implant) to realign the medial edge and mobilize any remaining pliable tissue to improve the glottic sound source. The algorithm in Figure 2 includes an option for considering reduced pliability during ML. In the illustrative case mentioned previously, the patient’s vocal fold pliability was improved by removal of a previously injected material that was too superficial. Several weaknesses of this study are readily apparent. The retrospective design of this study presents a weakness that could be overcome by designing a hypothesis-driven prospective study that considered factors contributing to the need for revision medialization laryngoplasty. Complete pre- and postoperative voice measures for all patients within the cohort would potentially make conclusions more powerful. Patients in this study had variable follow-up due the nature of referral patterns from outside institutions and their ability to travel for repeat examinations. Adequate longterm follow-up for all patients would enhance the results.
Conclusions Patterns of failure in patients with suboptimal phonatory function after transcervical medialization laryngoplasty included persistent glottic incompetence, glottic overclosure,
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
6
Otolaryngology–Head and Neck Surgery
and decreased vocal fold pliability. Closure of a persistent posterior glottic gap and proper assessment of phonatory pliability were important factors for revision in this patient cohort. Revision transcervical medialization surgery, guided by individualized consideration of vocal fold position and surface pliability, can improve phonatory outcomes as assessed by subjective (VRQOL) as well as objective acoustic and aerodynamic measures of vocal function. Author Contributions Noah P. Parker, wrote article, collected data, analyzed data, revised article, final approval, accountable for all aspects of the work; Anca M. Barbu, analyzed revision data, revised article, final approval, accountable for all aspects of the work; Robert E. Hillman, collected voice data, analyzed voice data, revised article, final approval, accountable for all aspects of the work; Steven M. Zeitels, analyzed revision data, revised article, final approval, accountable for all aspects of the work; James A. Burns, designed study, analyzed revision data, revised article, final approval, accountable for all aspects of the work.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Payr E. Plastik am Schildknorpel zur Behebung der Folgen Einsetiger Stimmbandlahmung. Dtsch Med Wochensch. 1915; 43:1265-1270. 2. Isshiki N, Morita H, Okamura H, et al. Thyroplasty as a new phonosurgical technique. Acta Otolaryngol (Stockh). 1974;78: 451-457. 3. Koufman JA. Laryngoplasty for vocal cord medialization: an alternative to Teflon. Laryngoscope. 1986;96:726-730. 4. Montgomery WM, Blaugrund SM, Varvares MA. Thyroplasty: a new approach. Acta Otol Rhinol Laryngol. 1993;102:571-579. 5. Cummings CW, Purcell LL, Flint PW. Hydroxylapatite laryngeal implants for medialization: preliminary report. Ann Otol Rhinol Laryngol. 1993;102:843-851. 6. Hoffman HT, McCulloch TM. Anatomic considerations in the surgical treatment of unilateral laryngeal paralysis. Head Neck. 1996;18:174-187. 7. Isshiki N, Tanabe M, Sawada M. Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol. 1978;104:555-558. 8. Zeitels SM, Hochman I, Hillman RE. Adduction arytenopexy: a new procedure for paralytic dysphonia with implications for implant medialization. Ann Otol Rhinol Laryngol. 1998;107:2-24.
9. Zeitels SM. Adduction arytenopexy with medialization laryngoplasty and cricothyroid subluxation: a new approach to paralytic dysphonia. Op Tech Otolaryngol Head Neck Surg. 1999;10:9-16. 10. Zeitels SM, Hillman RE, Desloge RB, et al. Cricothyroid subluxation: a new innovation for enhancing the voice with laryngoplastic phonosurgery. Ann Otol Rhinol Laryngol. 1999;108:1126-1131. 11. Hogikian ND, Sethuraman G. Validation of an instrument to measure voice-related quality of life (VRQOL). J Voice. 1999; 13:557-569. 12. Koufman JA, Postma GN. Revision laryngoplasty. Op Tech Otolaryngol. 1999;10:61-65. 13. Maragos NE. Revision thyroplasty. Ann Otol Rhinol Laryngol. 2001;110:1087-1092. 14. Woo P, Pearl AW, Hsiung M, et al. Failed medialization laryngoplasty: management by revision surgery. Otolaryngol Head Neck Surg. 2001;124:615-621. 15. Anderson TD, Spiegel JR, Sataloff RT. Thyroplasty revisions: frequency and predictive factors. J Voice. 2003;17:442-448. 16. Cohen JT, Bates DD, Postma GN. Revision Gore-Tex medialization laryngoplasty. Otolaryngol Head Neck Surg. 2004;131: 236-240. 17. Andrews BT, Van Daele DJ, Karnell MP, et al. Evaluation of open approach and injection laryngoplasty in revision thyroplasty procedures. Otolaryngol Head Neck Surg. 2008;138:226-232. 18. Lundeberg MR, Flint PW, Purcell LL, et al. Revisions medialization thyroplasty with hydroxylapatite implant. Laryngoscope. 2001;121:999-1002. 19. Netterville JL, Stone RE, Luken ES, et al. Silastic medialization and arytenoid adduction: the Vanderbilt experience. Ann Otol Rhinol Laryngol. 1993;102:413-424. 20. Ustundag E, Boyaci Z, Keskin G, et al. Soft tissue response of the larynx to silicone, Gore-Tex, and irradiated cartilage implants. Laryngoscope. 2005;115:1009-1014. 21. Paniello RC, Dahm JD. Reversibility of medialization laryngoplasty: an experimental study. Ann Otol Rhinol Laryngol. 1997;106:902-908. 22. Duke SG, Salmon J, Blalock PD, et al. Fascia augmentation of the vocal fold: graft yield in the canine and preliminary clinical experience. Laryngoscope. 2001;111:759-764. 23. Hirano M. Clinical examination of the voice. In: Arnold GE, Winckle F, Wyke BD, eds. Disorders of Human Communication. New York, NY: Springer-Verlag; 1981:81-84. 24. Omori K, Slavit DH, Kacker A, et al. Quantitative criteria for predicting thyroplasty type 1 outcomes. Laryngoscope. 1996; 106:689-693.
Downloaded from oto.sagepub.com at UNIV NEBRASKA LIBRARIES on September 8, 2015
