Clinical Features and Surgical Outcomes Following Closed Reduction of Arytenoid Dislocation

Importance  Arytenoid dislocation is a rare condition characterized by vocal fold immobility and is easily mistaken as recurrent laryngeal nerve paralysis.

Objective  To describe the presenting features, multimodal diagnostic evaluation, and surgical outcomes following closed reduction (CR) of arytenoid dislocation.

Design, Setting, and Participants  Prospective case series at a single academic medical center. Evaluation and treatment data were obtained from 22 consecutive patients with arytenoid dislocation over a 7-year period.

Interventions  Patients underwent direct laryngoscopy and CR of the dislocated arytenoid, with adjunct injection laryngoplasty or botulinum toxin administration in select cases.

Main Outcomes and Measures  Initial diagnosis was confirmed using flexible laryngeal endoscopy with stroboscopy, computed tomography, electromyography, and interoperative palpation. Arytenoid motion (primary outcome measure) and vocal function data (secondary outcome measures) were collected before treatment and up to 6 months after treatment.

Results  Key history features included emergent intubation, elective intubation, and external laryngeal trauma. Sixteen patients (73%) had anterior and 6 patients (27%) posterior dislocation. One patient experienced spontaneous recovery. Following CR, with or without adjunct therapy, 18 of the remaining patients (86%) exhibited arytenoid motion recovery with concomitant voice improvement. Recovery was sustained at 6 months after CR. Closed reduction performed within 21 days of the presumed dislocation event was associated with a superior arytenoid motion recovery rate.

Conclusions and Relevance  These data represent the largest clinical series on arytenoid dislocation with complete vocal function data and follow-up at 6 months after CR. These findings also corroborate existing evidence for early surgical intervention.

Arytenoid dislocation, classically defined as complete separation of the arytenoid cartilage from the cricoarytenoid joint, is a rare event, with fewer than 150 cases reported in the clinical literature to date.^1-9 Although the overall incidence of arytenoid dislocation is unknown, it may be underdiagnosed and misidentified as recurrent laryngeal nerve (RLN) paralysis in certain patients.^2,6,10 Both arytenoid dislocation and RLN paralysis are characterized by arytenoid immobility resulting in dysphonia and potential dysphagia; however, accurate differential diagnosis is critical, since surgical management is disorder specific.^10,11 Existing evidence suggests that early identification and treatment is associated with a higher likelihood of arytenoid motion restoration and voice improvement.^2,3,8 Consequently, improved awareness and diagnostic algorithms among primary otolaryngologists and laryngologists are important to optimizing outcomes in this patient population.

Because of its presumed low incidence and the aforementioned diagnostic challenges, the majority of published data on arytenoid dislocation are in the form of retrospective case series, with small sample sizes.¹ These previous reports have variably relied on clinical history features, laryngeal endoscopy, computed tomography (CT), and electromyography (EMG); however, there remains no consensus on the most helpful diagnostic tools among these modalities.^1,2,6,10-12 Furthermore, no treatment studies have documented long-term stability of the repositioned arytenoid or quantitative vocal function change. Herein, in a 7-year prospective study at a single institution, we recruited 22 consecutive patients with arytenoid dislocation, collected comprehensive multimodal diagnostic data, performed direct laryngoscopy with closed reduction (CR), and evaluated arytenoid motion and vocal function outcomes up to 6 months after treatment.

This study was approved by the institutional review board of Soonchunhyang University School of Medicine. Twenty-two consecutive patients were enrolled in a prospective clinical protocol from February 2007 to July 2013 (Table 1). All patients provided written informed consent. A multimodal diagnostic workup was performed using a combination of clinical history taking, flexible endoscopy with stroboscopy (Figure 1A), thin-slice high-resolution laryngeal CT (Figure 1B and C), and laryngeal EMG (bilateral thyroarytenoid and cricothyroid muscles). The following clinical features were considered supportive of a positive arytenoid dislocation diagnosis: (1) history of a traumatic laryngeal event associated with sudden-onset dysphonia/dysphagia; (2) arytenoid positional displacement, arytenoid immobility with residual cuneiform motion,¹² and an absent jostle sign¹⁰ on laryngeal endoscopy; (3) arytenoid positional displacement or frank cricoarytenoid joint obliteration on laryngeal CT; and (4) normal motor unit recruitment on laryngeal EMG. Patients with EMG-confirmed RLN paralysis, other neurolaryngological disorders, or vocal fold mucosal disease were excluded from the study.

Laryngeal CT data were collected using a multidetector 16-row CT unit (Somatom Sensation 16; Siemens Medical Solutions). The scanning protocol involved collection of unenhanced and contrast-enhanced (Ultravist 370; Schering) CT images at 120 kVp and 60 to 270 mAs; field-of-view, 240 mm; collimation, 0.625 to 0.75 mm; pitch, 0.75 to 0.984; slice thickness, 2 mm; and reconstruction in coronal and axial planes. Adaptation of the tube current with respect to body mass was used as a standard dose-saving measure. All scans were interpreted by a board-certified radiologist with specialization in head and neck pathology.

Following a positive diagnosis, patients underwent direct laryngoscopy followed by CR, as described by Sataloff et al.¹⁰ Induction anesthesia was achieved using intravenous sedation and muscle relaxants. Arytenoid manipulation procedures were then performed during apneic anesthesia with intermittent ventilation.¹³ Immediately prior to CR, intraoperative arytenoid palpation was performed to confirm the diagnosis and rule out arytenoid fixation.

To reduce an anterior dislocation, a Hollinger laryngoscope was positioned against the medial surface of the anteromedially dislocated arytenoid. The laryngoscope was then rotated approximately 130° in the direction opposing the dislocated arytenoid, allowing the upper surface of the laryngoscope to make broad contact with the medial surface of the cartilage. The surgeon's contralateral hand was placed externally on the larynx to provide counterpressure. The arytenoid was forcibly manipulated between the laryngoscope tip and the fingers of the surgeon's hand to achieve reduction (Figure 2A).

To reduce a posterior arytenoid dislocation, the tip of a Miller 3 blade was placed against the posterior surface of the posterolaterally dislocated arytenoid. Once the cartilage was hooked with the lip of the blade, force was applied to provide cephalic arytenoid distraction, followed by anteromedial reduction (Figure 2B and C).

Adjunct injection laryngoplasty (n = 4) was performed to address residual posterior glottal insufficiency following CR for posterior arytenoid dislocation^10,14; adjunct posterior cricoarytenoid muscle botulinum toxin injection (n = 1) was performed to address intrinsic laryngeal muscle imbalance following CR for posterior arytenoid dislocation.²

Arytenoid position and motion were evaluated prior to CR, and at 1 and 2 weeks and 1 and 6 months after CR. One patient underwent an additional evaluation at 2 months after CR. Vocal function data were collected prior to CR and at 1 and 6 months after CR. Mean jitter, mean shimmer, and harmonics to noise ratio (HNR) data were extracted from the steady state portion of a sustained [a] vowel token using the Multi-Dimensional Voice Program (MDVP model 4500; Kay Pentax). Maximum phonation time (MPT), produced on a sustained [a] vowel, was recorded using the Computerized Speech Laboratory (CSL model 4500; Kay Pentax) and averaged across 3 trials. Psychosocial data were collected using the Korean language version of the Voice Handicap Index (VHI). Endoscopic and videostroboscopic data were collected using the Rhino-Laryngeal Stroboscope (RLS model 9100; Kay Pentax). Glottal gap was rated using a previously reported 4-point equal-appearing interval scale (0, severe; 1, moderate; 2, mild; and 3, absent).¹⁵ Mucosal wave was rated using a previously reported 4-point equal-appearing interval scale (0, absent; 1, severely reduced; 2, mildly reduced; and 3, intact).¹⁵

Acoustic, aerodynamic, psychosocial, and vocal fold physiologic data were analyzed using a Wilcoxon signed-rank test. Arytenoid motion recovery rates were compared for patients undergoing early (within 21 days of presumed dislocation event) vs late (beyond 21 days of presumed dislocation event) CR using Kaplan-Meier estimates and a log-rank test. A complementary analysis of final arytenoid motion outcome at 6 months after CR was performed using the Fisher exact test. A type I error rate of .05 was used for all analyses. All P values were 2 sided.

The presumed etiology of arytenoid dislocation was emergent intubation in 8 patients (36%), planned endotracheal intubation during general anesthesia in 8 patients (36%), and external laryngeal trauma in 6 patients (27%) (Table 1). No patients had a recent history of thoracic or anterior neck surgery. Using our diagnostic protocol, we diagnosed 16 patients (73%) as having anterior dislocation and 6 patients (27%) as having posterior dislocation. All patients underwent flexible endoscopy with stroboscopy, laryngeal CT, and interoperative palpation; 17 patients (77%) underwent laryngeal EMG. Laryngeal CT resulted in the radiologist offering a suggested diagnosis of arytenoid dislocation in a single patient (5%), RLN paralysis in 15 patients (68%), and no pathologic finding in 6 patients (27%). Of the 17 patients who underwent laryngeal EMG, 15 (88%) showed intact neuromuscular function, whereas 2 (12%) exhibited reduced motor unit recruitment in the thyroarytenoid muscle ipsilateral to the dislocated arytenoid.

Twenty-one patients underwent direct laryngoscopy with CR. Of these, 4 patients underwent adjunct injection laryngoplasty and 1 patient underwent adjunct botulinum toxin injection to the posterior cricoarytenoid muscle. One patient demonstrated spontaneous recovery and was therefore not scheduled for surgical intervention. This patient’s data were removed from all clinical outcome analyses. Following CR, 18 patients (86%) experienced immediate arytenoid motion recovery that was sustained for at least 6 months. No serious adverse events, such as airway obstruction, were observed following CR; however, 1 patient developed a minor vocal fold hematoma that resolved spontaneously without additional treatment. Acoustic, aerodynamic, psychosocial and vocal fold physiologic measures uniformly showed significant improvement at 1 and 6 months after CR (P < .05) (Figure 3A).

The interval from presumed arytenoid dislocation event to surgical intervention ranged from 7 to 6223 days (approximately 17 years) with a median of 21 days. Thirteen patients (64%) underwent CR within 21 days of their presumed dislocation event. Analysis of Kaplan-Meir estimates showed a significantly increased rate of arytenoid motion recovery in these patients compared with those who underwent CR beyond 21 days of their presumed dislocation event (P < .001) (Figure 3B). This finding was confirmed by a complementary analysis of final arytenoid motion status 6 months after CR (P = .04) (Table 2).

Arytenoid dislocation was first reported in 1973¹⁶ and since that time has been described in various case series, most of which have been retrospective and characterized by small sample sizes, varying approaches to diagnosis, varying duration of clinical follow-up, and varying documentation of posttreatment arytenoid position, arytenoid motion, and vocal function.¹ In the present prospective observational study from a single institution, we evaluated etiological and presenting clinical features in 22 consecutive patients; collected a complete diagnostic array of endoscopic, EMG, and CT data in the majority of these patients; treated the patients with direct laryngoscopy and CR; and evaluated arytenoid motion and vocal function outcomes up to 6 months after treatment. We observed a uniform history of intubation-related or external laryngeal trauma; a higher incidence of anterior (vs posterior) dislocation; sustained arytenoid motion recovery and voice improvement following CR; 1 case of spontaneous recovery; and a superior arytenoid motion recovery rate when CR was performed within 21 days of the presumed dislocation event. To our knowledge, this data set represents the largest clinical series on arytenoid dislocation with complete vocal function data and follow-up at 6 months after CR. Furthermore, these findings corroborate existing evidence for the importance of early surgical intervention.^2,3,8

The majority of patients in our study (73%) experienced a presumed inciting event that occurred during an emergent intubation or planned endotracheal intubation under general anesthesia. This observation is consistent with prior reports of arytenoid dislocation following endotracheal intubation, direct laryngoscopy, or blunt laryngeal trauma, and in some cases following laryngeal mask airway placement, esophagoscopy, or gastroscopy.^1,7,11,17-19 The estimated incidence of arytenoid dislocation following intubation is 0.02% to 0.1%^7,18,19; however, the overall incidence of this disorder is unknown. Arytenoid dislocation is likely underidentified in the treatment-seeking population, in part owing to its clinical similarity to RLN paralysis.^2,4,6,10 Both disorders result in vocal fold immobility, so unless there is a strong clinical suspicion of cricoarytenoid joint trauma and/or EMG evidence of intact RLN function, arytenoid dislocation may not be considered as a probable diagnosis at initial presentation. Of note, arytenoid malrotation may also occur in the congenitally asymmetric larynx²⁰: as such, this should be considered in the differential diagnosis.

Arytenoid dislocation is generally diagnosed via exclusion of other conditions, particularly RLN paralysis; however, there is no firm consensus on the most appropriate or efficient diagnostic workflow when evaluating for this disorder.¹ In the present study, we provisionally diagnosed arytenoid dislocation based on clinical history features, flexible endoscopy, laryngeal CT, and laryngeal EMG. Confirmation of the diagnosis was obtained using interoperative palpation. It is clear from both our data and earlier reports that individual elements of this diagnostic array have poor sensitivity when considered in isolation. For example, despite using thin-slice high-resolution CT, radiologic imaging alone accurately identified arytenoid dislocation in a single case, yielding sensitivity of just 5%. Of note however, the radiologist’s suspicion of arytenoid positional impairment (ie, dislocation or RLN paralysis) occurred in 73% of patients, similar to the laryngeal CT-based identification rates for arytenoid dislocation reported in prior retrospective studies^2,21 and highlighting the challenge of separating dislocation from paralysis using radiologic imaging only. Superior diagnostic outcomes might be obtained when using a smaller field of view that is focused on the glottis and cricoarytenoid joint, a slice thickness less than the 2 mm used here, as well as 3-dimensional image rendering.^1,6 Furthermore, although negative EMG findings are considered a hallmark of arytenoid dislocation, we observed reduced thyroarytenoid motor unit recruitment in 2 patients. This observation has been made in previous studies and suggests concomitant neuropathy and/or postinjury muscle fibrosis.^2,4,10 Overall, it appears that high diagnostic confidence requires identifying arytenoid immobility and positional asymmetry, subtle dynamic features (residual cuneiform motion¹² and an absent jostle sign¹⁰), normal motor unit recruitment on laryngeal EMG (in most, but not all, cases), and evidence of dislocation on palpation, particularly in patients with a positive history of recent intubation or external laryngeal trauma. Given the relationship between early surgical intervention and favorable clinical outcomes, seen in both our data set and earlier reports,^2,3,8 it is imperative that patients with a clinical history suggesting possible cricoarytenoid joint trauma undergo a rapid and multifaceted workup.

We observed statistically significant improvement in arytenoid motion as well as acoustic, aerodynamic, psychosocial, and vocal fold physiologic function that was sustained at 6 months after CR. Of note, 1 patient experienced spontaneous resolution of symptoms, which in the literature has been attributed to arytenoid repositioning during forceful laryngeal maneuvers such as coughing or emesis.^10,22 All 13 patients who underwent CR within 21 days of their presumed dislocation event experienced complete return of arytenoid motion, whereas 3 of 8 patients who underwent CR beyond 21 days of their presumed dislocation event had no return of function at 6 months after CR. These 3 patients were the only individuals in the study with an interval from presumed dislocation to attempted CR of more than 1 year; in 1 case the interval was approximately 17 years. Early intervention is believed to result in superior outcomes because of the limited opportunity for cricoarytenoid joint ankylosis to develop.^2,10 We performed adjunct injection medialization in 4 cases to address residual posterior glottal insufficiency following CR for posterior arytenoid dislocation and delivered adjunct botulinum toxin to the posterior cricoarytenoid muscle in 1 case of posterior arytenoid dislocation to assist restoration of intrinsic laryngeal muscle balance. Use of these adjunct therapies was based entirely on clinical presentation at the time of CR and was not considered in our analysis of treatment outcomes.

Our data highlight the importance of identifying intubation-related or external laryngeal trauma events when taking a clinical history and support the importance of expedient multimodal workup on suspicion of arytenoid dislocation. Direct laryngoscopy with CR results in restoration of arytenoid motion and vocal function that is sustained for at least 6 months. Early diagnosis followed by CR within 21 days of dislocation is associated with a superior rate of arytenoid motion restoration.

Submitted for Publication: April 17, 2014; final revision received July 16, 2014; accepted August 4, 2014.

Corresponding Author: Nathan V. Welham, PhD, Division of Otolaryngology–Head and Neck Surgery, Department of Surgery, University of Wisconsin School of Medicine and Public Health, 600 Highland Ave, K4/723 CSC, Madison, WI 53792 (welham@surgery.wisc.edu).

Published Online: September 25, 2014. doi:10.1001/jamaoto.2014.2060.

Author Contributions: Drs Lee and Welham had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: Welham.

Statistical analysis: Welham.

Obtained funding: Welham.

Administrative, technical, or material support: All authors.

Study supervision:Welham.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by grants R01 DC004428 and R01 DC010777 from the National Institute on Deafness and other Communication Disorders and a grant from the Soonchunhyang University Research Fund.

Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Previous Presentation: This study was presented at the 135th Annual Meeting of the American Laryngological Association; May 15, 2014; Las Vegas, Nevada.

Additional Contributions: Glen E. Leverson, PhD, Department of Surgery, University of Wisconsin School of Medicine and Public Health, Madison, provided statistical consultation. No financial compensation was provided.