Systematic Review of Laryngeal Fractures and Trends in Operative Management

Abstract

Study Design: Systematic review of the literature. Objective: The goal of this study is to review the current literature on the trends in management of laryngeal fractures following trauma. Methods: Independent searches of the PubMed and MEDLINE databases were performed. Articles from the period of 1963 to 2020 were collected. All studies which described laryngeal fractures using the Boolean method and relevant search term combinations, including “Laryngeal”, “Fracture”, “Operative”, and “Management” were collected. Results: A total of 588 relevant unique articles were identified for analysis. Of these, 24 articles were deemed appropriate for inclusion in the literature review. Due to variability in study design and outcome measures, formal synthesis of data in the form of a meta-analysis was not possible. Conclusions: Laryngeal fractures are rare traumatic injuries that require early identification and evaluation with complex management options. This comprehensive review aims to highlight the breadth of the topic with regard to presentation and clinical management. Though there remains no clear best practice for laryngeal fracture management, we review trends in clinical practice throughout the literature.

Introduction

Laryngeal trauma is a rare yet potentially devastating injury which presents a complex diagnostic and therapeutic challenge to even the most experienced craniomaxillofacial trauma surgeon. There is an estimated incidence of 1 in 30,000 emergency department visits in the United States [1]. Timely identification by emergency physicians and otolaryngologists alike, as well as expedited management of acquired laryngeal framework injuries is paramount in the care of these injuries as mortality related to blunt and penetrating mechanisms ranges from 20 to 40% [2,3,4]. As concomitant airway injury is often associated with multisystem organ trauma, patients are frequently triaged and systematically assessed according to basic principles set forth by American College of Surgeons Advanced Trauma Life Support protocol [5].

Laryngeal fractures are known to adversely affect all native functions of the larynx. Significant injury should be suspected in patients with acute symptoms of dysphagia, odynophagia, hoarseness, hemoptysis, and respiratory distress. Physical examination findings of concern include subcutaneous emphysema, change of laryngeal framework contour, and the presence of anterior cervical ecchymosis or lacerations, although it is not uncommon for patients to initially present with a benign airway examination. Typically, ancillary imaging may be performed in stable patients to establish the extent and location of injury and provide objective clinical information for definitive treatment planning.

Initial management of severe laryngeal trauma requires establishment of a secure airway either in the form of endotracheal intubation under direct fiberoptic visualization or tracheotomy for patients with signs of impending airway compromise. The laryngeal framework can then be thoroughly assessed for integrity and need for potential surgical exploration. Proper restoration of the dysfunctional larynx requires early and precise repair of the complex cartilaginous architecture. Most reports recommend exploration and reconstruction of the laryngeal framework within 48 hours [1,6]. Failure to properly address injuries in a timely fashion may lead to permanent dysphonia and chronic airway obstruction.

Despite clinical advances in the operative management of laryngeal trauma over the last two decades, there currently remains a lack of consensus in the surgical literature with regards to optimal methods for functional repair of the laryngeal framework. In this systematic review, we investigate various practices surrounding identification, classification, and management of laryngeal fractures and explore possible predictive factors for optimal surgical techniques and clinical outcomes.

Materials and Methods

Study Protocol

This study was designed and performed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Protocol [7,8] and was deemed exempt from institutional review board approval. Independent searches of the PubMed and MEDLINE databases were performed from November 23, 2017 to February 1, 2020 by two authors (A.E.M and V.A.P) to identify studies which described the clinical manifestations, diagnostic principles, and management of traumatic laryngeal fractures. English and full-length original articles were assessed for data extraction eligibility. Articles from the period of 1963 to 2020 were collected. All studies which described laryngeal fractures using the Boolean method and relevant search term combinations, including “Laryngeal”, “Fracture”, “Operative”, and “Management” were collected. References of included articles were also reviewed for eligibility. This material was supplemented with targeted searches to address specific needs identified in writing this comprehensive review and meta-analysis. The utilized search strategy is summarized in Figure 1.

Data Collection and Synthesis

Studies were categorized by lead author, year and country of publication, and type of study. Each article was assessed using the methodological index for non-randomized studies (MINORS) instrument [9]. Baseline characteristics included study sample size, mean age at presentation, gender, mechanism of injury, Schaefer-Fuhrman classification of injury, injury site, fracture pattern, laterality, ancillary imaging, and technique utilized for laryngeal fracture repair [10,11]. In addition, presenting symptoms were also analyzed given the availability of published clinical data. This included pain, dysphagia, odynophagia, voice changes, hemoptysis, respiratory distress, subcutaneous emphysema, and loss of consciousness. The primary outcomes with regards to symptom resolution included vocal quality, decannulation status, and swallowing difficulties. Missing data were excluded from the quantitative synthesis.

Statistical Analyses

Descriptive statistical analyses were performed using SPSS statistical software (IBM) on Windows 7. Formal statistical analysis was precluded by the heterogeneity of the studies.

Results

A total of 588 articles were identified upon initial search. After duplicate removal and review for eligibility criteria, 24 studies from 10 countries were deemed eligible for data extraction. As depicted in

Table 1. Studies Analyzed.

, publication dates ranged from 1963 to 2019. There were 4134 patients included with laryngeal fracture having a mean age of 34 years. As shown in

Table 2. Characteristics of Laryngeal Fracture.

, gender was reported on 4092 patients, 84.8% of which were male (3469). Mechanism of injury was reported in 4130 patients, the majority (85.9%) of which were blunt (3547), with the remaining 14.1% penetrating.

Patients presented with a variety of symptoms including change in voice (41.3%), respiratory distress (16.0%), pain (11.5%), subcutaneous emphysema (9.2%), and dysphagia (8.9%). A total of 537 patients underwent imaging with CT, while a single patient had ultrasound performed. Site of injury included primarily thyroid cartilage (72.1%) and cricoid cartilage (24.6%), with the Schaefer-Fuhrman injury group for most patients being II (32.9%) or III (32.3%). Reported fracture pattern was primarily non-comminuted (75.8%), with 14.1% of patients experiencing laryngotracheal separation and 10.1% of patients having comminuted fractures. Reported laterality of fractures was mostly unilateral (82.5%). A total of 335 patients were reported to have tracheostomy, with 79 of these patients (23.6%) reported to be decannulated over an average of 10.4 days.

As demonstrated in

Table 3. Interventions and Outcomes of Laryngeal Fracture. Variable Patients (N = 4134).

, specific therapeutic intervention was only commented on for 618 patients. Just over half of those patients underwent some kind of operative intervention, while the remaining 41.5% reported were observed. Interventions performed included plate fixation using titanium (18.6%) and bioabsorbable (0.8%), suture fixation (26.0%), stenting (9.2%), and laryngotracheal anastomosis (2.4%). Postoperatively, half of patients reported voice quality as good (54.4%), with an average voice handicap index (VHI) of 4.5. Vocal fold paralysis was reported in 19 patients, including 4 patients with bilateral paralysis. Swallowing was reported to be adequate for the majority of patients (92.3%).

Discussion

Laryngeal framework injuries present complex diagnostic and therapeutic challenges. We sought to assess the current literature to better define methodology for functional repair of these injuries.

Demographics

Laryngeal fracture most often occurs in young male patients. The average age of the injured patient included in our review was 34 and 85% of patients were male, consistent with data for general trauma patients [12]. Most patients sustain injury from blunt trauma; we demonstrate 86% of patients suffering blunt trauma, with most resultant from motor vehicle collision, assault, or athletics [13]. A high index of suspicion for laryngeal injury should be had in any patient with blunt injury to the neck, and, depending upon the force involved, the laryngeal injury sustained in blunt trauma may range from isolated cartilaginous fracture to complete laryngotracheal separation. In the assessment and management of laryngeal fracture, mechanism of injury including energy of impact and trajectory, particularly in the case of penetrating injury, can provide important insight to development of airway compromise.

Clinical Manifestations of Laryngeal Fracture

A number of clinical symptoms have been shown to be associated with these fractures. We found voice changes to be the overwhelming primary presenting symptom, occurring in 41.3% of cases. Additionally, respiratory distress (16.0%) and pain (11.5%) were often presenting symptoms. These initial symptoms particularly in the clinical setting of traumatic injury should raise suspicion for laryngeal fracture which in turn can help the clinician to prepare for developing complications. On physical examination, the presence of hemoptysis or crepitus should raise suspicion for laryngeal fracture; though less common than other presenting symptoms, these findings were present in 6.1% and 9.2%, respectively.

Assessment of Potential Laryngeal Injury

Evaluation of laryngeal injury requires the clinician first ensures a patent airway. When airway stability is not of imminent concern, extracting details regarding mechanism of injury and temporal presentation should be priority in patient discussion. Physical examination may yield evidence of more than laryngeal injury. The astute clinician should assess for concurrent injury to neurovascular structures as well as for signs of pharyngeal or esophageal injury [14]. Furthermore, the status of the cervical spine is of the utmost concern when manipulating the patient for nasopharyngeal laryngoscopy (NPL) or, if necessary, intubation.

One of the most important aspects of clinical evaluation is the flexible NPL, which is performed to best assess airway patency, vocal fold mobility, and mucosal integrity throughout the upper airway. Although NPL provides important information related to patency and airway integrity, high-resolution fine-cut CT remains the gold standard for assessment of the injured airway, particularly in the setting of blunt injury when edema is noted on physical exam. The decision to pursue CT must include consideration of risks and benefits associated with the current airway status. An unsafe airway or open injury mandating neck exploration should be managed prior to diagnostic intervention. While CT is preferred, ultrasound was used to evaluate one patient in our review. Focused airway ultrasound has been suggested to play a role in the diagnosis of laryngeal fracture, while expediting the process [15].

Classification Groupings of Laryngeal Injury

Laryngeal injuries are classified by severity, most widely using a scale proposed by Fuhrman and Schaefer [10,11]. Group I injuries represent a stable airway with minor endolaryngeal hematoma or laceration. Group II suggest airway compromise, including more severe injury to the soft tissue or a single nondisplaced laryngeal fracture. Group III describe injuries of massive edema, cartilage exposure, displaced fracture, or vocal cord immobility. Group IV represents unstable comminuted laryngeal fracture, while Group V is the most severe demonstrating complete laryngotracheal separation. Conservative management is recommended for Groups I and II, which has been found to be 100% effective in this population [6]. Surgical intervention is necessary for Groups III–V.

Earlier schema were further contemporized to the Legacy Emanuel Hospital and Health Center (LEHHC) classification by Verschueren and colleagues to include more advanced technological findings and further guide management technique. They instead create four stages where each progressive stage adds additional management [16]. Another classification schema commonly refers to anatomical location of injury, primarily supraglottic, glottic, and subglottic injury [17].

The Schaefer-Fuhrman classification was only ascertained for 155 patients in this review, many of which were not reported in the published article and instead required retrospective review. Of these, 36.8% could be classified as Groups I and II with possible consideration of conservative management, and 10.3% were Group V, complete separation. The remaining patients fell into the middle of the grading scale.

Despite protocols and classification schema, many patients are excluded, as these primarily address patients presenting immediately following trauma. Those without emergent airway compromise often present days to weeks following injury. Furthermore, we noted that few of our identified studies ultimately utilized this classification at the time of publishing. Instead, classification for the purposes of this study required retrospective categorization. While formerly this rare diagnosis relied heavily on clinical acumen, with modern updates to these classification guidelines, including both diverse diagnostic data and management guidance, the literature may move toward support of a more universal guide for treatment modality.

When assessing injury by most common site, the thyroid cartilage accounted for 72.1% of injuries. Fracture to the thyroid cartilage is known to cause vocal cord edema and resultant voice changes. Cricoid fracture was also commonly reported in 24.6% of cases. Of note, fracture to the cricoid places the patient at great risk for airway compromise as the circumferential structure of the cartilage can quickly become edematous requiring rapid intervention and airway stabilization [17].

Management of Laryngeal Fracture

The principles for managing laryngeal fracture are similar to those of any fracture, namely, restoration as close to normal anatomical and functional positioning as possible, which in the larynx involves the preservation of phonation, respiration, and deglutition. Given the spectrum of injury, amount of cartilaginous displacement, and airway stability, management may be surgical or nonsurgical.

Schaefer and colleagues have put forth a diagnostic algorithm to guide initial management for these patients. Intubation should be considered when the following three criteria are met: (1) the larynx and trachea are in continuity and clearly intact, (2) the airway is visible to direct inspection by endoscopy, and (3) a highly experienced physician is available to perform the intubation. If any of these criteria are unmet, management should proceed with cricothyrotomy or tracheotomy [2]. We found a total of 335 tracheostomies reported. In the hands of experts, mild to moderate injuries can be repaired without tracheostomy; however, severe fractures or separation warrant a tracheostomy.

Once the airway is stabilized, NPL examination should be performed to ensure anatomical placement and review need for additional management. With appropriate workup, many patients may not require surgical intervention, as was the finding with 41.5% of reported cases. This recommendation can be safely followed for patients in Schaefer Groups I and II, which accounted for 36.8% of our reported population. Observation for up to 2 days immediately following injury to ensure continued airway stability can be considered, while nonsurgical interventions are utilized, commonly including humidified air, head of bed elevation, and voice rest. Airway edema in these patients may be managed using steroids and a prophylactic proton pump inhibitor to minimize laryngopharyngeal reflux. Serial NPL examinations supplemented by pulse oximetry is warranted during the immediate post-trauma period.

Nearly 60% of patients with laryngeal fracture underwent operative intervention, though the method of fixation for laryngeal fracture has evolved over the past decades. Historically, wire fixation techniques were favored, though this has progressed to preferential utilization of suture fixation. We report 26% of operative interventions utilized suture fixation. The challenges associated with fixation alone are inherent to the natural structure and function of the larynx. Mobility of the fixed segments can alter an otherwise adequate reduction yielding tissue necrosis.

In order to minimize disruption to the fracture reduction by natural movement of the larynx, both endolaryngeal stenting and external fixation may be necessary for immobilization. When stenting is necessary, soft stenting is preferred for as minimal time as possible, likely weeks rather than months, however with increased risk for infection and granulation formation, the role of stenting has been disputed as of late [18,19]. Our review suggests a stenting rate of 9.2% when assessing all operative interventions. Stenting was initially recommended in early literature when surgical options were limited; however, even in the 1980s, it was recognized to be inhibitive to adequate healing with recommendations to use for no more than a few weeks [18]. More recent literature by Schaefer and colleagues suggests a role for reapproximation of the anterior commissure in order to limit the need for stent placement in instances where the fracture disrupts the native commissure [2].

To address the challenges associated with laryngeal fixation, in 1990 Woo first utilized miniplates for laryngeal framework reconstruction [20]. Although placement can be challenging, the rigid stability of the miniplate supports adequate alignment of the healing cartilage. Principles of plating are similar to that of bony defects; however, the soft cartilage poses challenges, namely, stripping of threads during attempted screw placement [21]. Some recommend using a drill bit smaller than the intended screw to prevent this difficulty [21]. Biodegradable plates have been proposed as an alternative to metal plates and resorb in 1 to 3 years limiting risk of migration and extrusion [22,23]. A number of advantages of the bioresorbable plate include their malleability to conform shape of the larynx while maintaining strength for 6 to 12 months. Additionally, bioresorbable plates in the larynx, as in other areas, carry a lower infection risk [23]. We report a 18.6% usage of titanium plates with only 0.8% utilization of bioabsorbable plates. Regardless of plate selection, plating using 4-point fixation creates alignment and limits the need for endolaryngeal stenting, an advantage over the 2-plane fixation present in suture reduction. Despite the improvements of plate fixation, stenting may be necessary in the repair of severely comminuted fractures, which we found to occur in 10% of reported cases.

Laryngotracheal separation is a rare occurrence, took place in 15 cases for a rate of 2.4% of all operative interventions. Repair in this setting commonly utilizes tracheal traction sutures to prevent retraction during primary anastomosis. Though the time to surgical repair was not measured in this review, recent literature suggest early operative intervention yields improved postoperative laryngeal function [1,6,18,24,25,26].

Outcomes of Laryngeal Fracture

Overall outcomes assessed in this population were poorly described. We report 54.4% of patients experienced good postoperative voice quality; however, a median voice handicap index of 4.5 supports markedly improved quality of vocal function. We report only 19 cases (<0.5%) of postoperative vocal fold paralysis, 4 of which were bilateral. Furthermore, postoperative swallowing was found to be of good functioning in 92.3% of patients.

Initial literature on the topic was based upon single surgeon experience, Schaefer, who reports on nearly three decades of work. Although much is discussed regarding mechanism of injury, no difference in functional outcomes is reported between blunt and penetrating trauma [1]. A 1986 review found poor outcomes with up to 40% of patients experiencing voice changes; however, more modern literature, including our own review, demonstrates significant improvement in these results [6,27].

Although we report only 335 cases of tracheostomy and 79 decannulations, it has been reported in the literature that nearly all patients will be decannulated with functional speech and swallow outcomes [6]. Of those reported, most patients appear to decannulate prior to hospital discharge. We report an average length of time to decannulation was 10.4 days, with a similar average length of stay at 10.8 days. As most tracheostomies are placed for acute airway patency and operative intervention, we suspect this data was largely underestimating the actual frequency of both tracheostomy and subsequent decannulation in our review.

Limitations

This review was largely limited by the heterogeneity of the reported studies, such that more formal meta-analysis was unable to be performed. While our data capture over five decades of literature on the topic, there was great variability of the common practice at the time each study was performed. Furthermore, many studies did not explicitly share all data points of interest in our review, and most notably lacking was the limited findings reported on flexible NPL examination.

Conclusions

Though uncommon, laryngeal fractures require early identification and intervention. There remains no clear best practice for the management of laryngeal fractures. Clinical trends suggest suspicion for fracture in young, male patients with blunt injury presenting with new onset voice changes. Assessment using NPL is vital to assess function, in addition to CT for anatomical evaluation. The Schaefer-Fuhrman notation has been most utilized in the literature to group injury in the acute setting, but does little to predict operative intervention and outcome. The incorporation of functional and anatomical data upon presentation will be useful to allow for future robust statistical analysis of protocols and outcomes. Based on our observations, we recommend further comprehensive studies to develop a universal guide to treatment modality.