Arytenoid cartilage necrosis due to prolonged endotracheal intubation: A case report of conservative follow-up with variable endoscopic findings

Abstract

Arytenoid cartilage necrosis (ACN) is a life-threatening laryngeal complication that develops after prolonged endotracheal intubation (PEI). We describe a case of ACN due to PEI that was treated conservatively with close endoscopic follow-up. A 46-year-old man required 4 days of postoperative endotracheal intubation after surgery for Stanford A acute aortic dissection. Five days post-extubation, he presented with hoarseness and wheezing. Endoscopy and computed tomography findings confirmed ACN. We used conservative treatment to avoid a tracheostomy because the patient had no dyspnoea. During healing, vocal cord mobility changed per the extent of cicatrisation and adhesion, and glottic closure due to inter arytenoid adhesion was possible. The bilateral vocal cords were eventually fixed in a paramedian position. The patient could phonate via ventricular band mobility without aspiration or dyspnoea. Given the possibility of variable laryngeal findings, careful endoscopic evaluations may be needed until wound healing completes to avoid sudden suffocation after PEI.

Keywords:

• Arytenoid cartilage necrosis
• prolonged endotracheal intubation
• laryngeal injury
• vocal cord mobility

Introduction

Prolonged endotracheal intubation (PEI) can cause varying degrees of laryngeal injury, including oedema, ulcer, granuloma, vocal cord (VC) immobility, and tracheal stenosis [1,2]. The optimal timing of tracheostomy has been discussed in numerous studies; for example, Mehel et al. [1] concluded that a tracheostomy should be performed for intubations that extend beyond 7 days.

The arytenoid cartilage (AC) can also be damaged by PEI, and complications such as mucosal oedema and dislocation of the AC have been reported [2]. However, few reports exist on AC necrosis (ACN), particularly reports that follow the healing process using endoscopy with a focus on VC mobility [2]. In this report, we present a case of ACN due to PEI that was treated conservatively with endoscopic follow-up, and we describe the variable endoscopic findings.

Case report

A 46-year-old man presented with hoarseness and a slight wheeze 5 days after extubation. He had undergone surgery for a Stanford A acute aortic dissection and required 4 days of postoperative endotracheal intubation with an 8-mm internal diameter endotracheal tube. He did not have any medical comorbidities except for that vascular disease and hypertension. His height, weight, and body mass index were 169 cm, 65 kg, and 22.8 respectively. The patient’s maximum phonation time was 4 s, which suggested laryngeal pathology (in the absence of pathology, many healthy adults can phonate beyond 20 s) [3]. Necrotic tissue and an ulcer were observed endoscopically on the surface of the bilateral posterior glottic region, including the AC. The VCs had lost almost complete mobility (Figure 1). The larynx had diffuse inflammation with redness and oedema. Computed tomography (CT) imaging showed a partially missing AC with a peripheral abscess. We diagnosed the patient with ACN (Figure 2). However, we did not perform a tracheostomy because the posterior glottic chink was fortunately confirmed and percutaneous oxygen saturation remained high without supplementation.

Figure 1. Endoscopic imaging findings at the initial visit. (A) The bilateral vocal cords have oedema and redness, and necrotic tissue with an ulcer is on the surface of the bilateral posterior glottic region. (B) A portion of the left arytenoid cartilage is exposed (surrounded by arrowheads).

Figure 2. Computed tomography findings. (A) An abscess is in the posterior glottic region (arrowhead). (B) A portion of the left arytenoid cartilage has disappeared (arrowhead).

Bacterial culture of the pus adhering to the AC revealed no heterogeneous bacteria such as methicillin‐resistant Staphylococcus aureus or Pseudomonas aeruginosa; however, the patient received treatment with intravenous antibiotics and did not receive any other treatment by surgery or medicine, such as corticosteroids or proton pump inhibitors. One week after the initial visit to our department, endoscopy revealed that the laryngeal inflammation had improved slightly. Granuloma growth was observed on the surface of the AC, although the necrotic tissue remained. Moreover, the right VC regained slight mobility, whereas the left VC did not.

Swallowing videofluorography, using a 1% thickened x-ray contrast medium, revealed no inflow to the larynx or pooling at the bilateral piriform sinus of the hypopharynx. The medium was washed out by swallowing multiple times. Videoendoscopic evaluation of swallowing showed good clearance with multiple swallows, no aspiration of 1% thickened water, and normal contact between the base of the tongue and the posterior pharyngeal wall. Based on these results, the patient began eating jelly-like food after starting dysphagia rehabilitation. The consistency of the food was gradually increased to solids, based on the findings from the videoendoscopic evaluation of swallowing. Two weeks after the initial visit to our department, endoscopic findings revealed mobility of the right VC but no mobility of the left VC (Figure 3); however, the movement of the right-side VC was not smooth. The patient was able to eat general foods without aspiration, although he needed a 1% thickened agent to drink, and was discharged.

Figure 3. Endoscopic imaging findings 2 weeks after the initial visit. (A) No oedema or redness is on the bilateral vocal cords. Necrotic tissue is slightly visible in the posterior glottic region. (B) The right vocal cord has active mobility on phonation.

One month after discharge (i.e. 7 weeks after the initial visit to our department), we endoscopically observed a smooth surface with no necrotic tissue in the posterior glottic region (Figure 4). However, the right VC lost complete mobility again, and the bilateral VCs were fixed in a paramedian position. The patient could phonate via the mobility of the ventricular bands but with hoarseness. At the last follow-up, no aspiration or dyspnoea was detected.

Figure 4. Endoscopic imaging findings 7 weeks after the initial visit. Bilateral vocal cords are completely fixed and the glottic space is present. No necrotic tissue is visible.

Discussion

Laryngeal injuries such as ulcers, oedema, VC immobility, tracheal stenosis, and dislocation of the AC are life-threatening complications caused by PEI [2]. These complications can develop because of persistent pressure against the laryngeal tissue from the endotracheal tube and cuff [4]. A tracheostomy is often required to prevent these complications in patients with PEI. Numerous studies have reported the optimal timing of conversion to tracheostomy [1]. However, an insufficient number of reports exist regarding conservative follow-up, based on endoscopic image findings in patients with ACN due to PEI.

In this patient, we initially suspected left recurrent laryngeal nerve paralysis because the patient had undergone post-vascular surgery for acute aortic dissection. However, endoscopic findings showed bilateral VC immobility and necrotic tissue attached to the posterior glottic region, including the AC. CT results confirmed the diagnosis of ACN. We did not perform a tracheostomy because the airway was open, and the patient was not dyspnoeic; however, the variable extent of adhesions required appropriate evaluation. During the healing process, we followed the patient carefully because of the possibility of sudden suffocation caused by airway obstruction due to adhesion in the posterior glottic region, including the AC. Therefore, if possible, phonation, swallowing, and dysphagia rehabilitation are not limited to the prevention of inter arytenoid adhesion.

The range of VC mobility was also variable in this patient. The cricoarytenoid joint has an important role in VC rotational movement [5]. We assumed that the bilateral VC immobility noted at the first visit developed because of the loss of the AC and VC interlocking movement around the cricoarytenoid joint caused by ACN. However, the right VC movement temporarily recovered. This recovery may have been induced by cicatrisation during healing, thereby forming a pseudo-linkage in place of the partial AC deficit. VC movement would finally be fixed when wound healing was completely achieved. This result showed that VC mobility depended on the extent of cicatrisation and inter arytenoid adhesion.

In a previous report, the division of an inter arytenoid adhesion was achieved using microlaryngoscopy with a laser or microsurgery scissors [2]. This procedure was avoided in our patient because we did not observe the closure of the glottic space with dyspnoea. Thus, after a laryngeal injury, VC mobility may vary in accordance with the extent of cicatrisation and adhesion that occurs during the healing process.

Colton et al. found no significant relationship between laryngeal injury, intubation period, and endotracheal tube size [6]. However, endoscopic evaluation for laryngeal injury may be required for all patients after PEI, regardless of these factors. Of note, sudden suffocation may be induced by the closure of the glottic space due to adhesion during the healing process, even if objective findings such as hoarseness and dyspnoea temporarily improve. To date, no reports exist that are well-controlled regarding the long-term conservative follow-up of cases of ACN due to PEI.

The main limitation of any case report is that it describes only one patient; thus, the results cannot be generalised. Therefore, further studies are required that focus on the endoscopic laryngeal findings during the healing process and that consider other factors such as age, sex, body mass index, and medical history to reveal whether conservative management without tracheostomy is effective for patients after PEI.

Conclusion

ACN is a life-threatening complication after PEI. We present a patient that was conservatively managed without tracheostomy; although careful continuous follow-up and appropriate endoscopic evaluations were needed, given the variable laryngeal findings during the healing process. Careful continuous follow-up and appropriate endoscopic evaluation are essential until wound healing is achieved to prevent sudden suffocation in patients after PEI.

Ethical statement

This study has not been published or presented elsewhere in part or in entirety. This material is the authors’ original work.

Patient consent

The patient described in the study provided written informed consent for the publication of this case report, and the study protocol was approved by the Ethics Committee of Nara Medical University Hospital (the proposal number was 3300).

Disclosure statement

The authors report no conflict of interest.

Data availability statement

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Additional information

Funding

This study did not receive any funding.