Introduction
The pathogenesis of coronavirus disease 2019 (Covid-19) is strongly associated with the upper airway. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes Covid-19, binds angiotensin-converting enzyme 2 (ACE2) and promotes cellular entry in association with host transmembrane serine protease 2 (TMPRSS2). Cells co-expressing ACE2 and TMPRSS2 have been identified within lung type II pneumocytes and nasal goblet secretory cells.Reference Ziegler, Allon, Nyquist, Mbano, Miao and Tzouanas1
Intubation of patients with Covid-19 poses a risk of viral transmission to healthcare workers, which can be extrapolated to the downsizing of endotracheal tube or reduced cuff pressures. Patients are therefore being intubated for prolonged periods with large cuff sizes. This increases the risk of subglottic or tracheal stenosis, and emergency re-intubation because of stenosis. Waiting 10 days or more before tracheostomy insertion increases the risk of subglottic stenosis.Reference Li, Yiu, Merrill, Yildiz, deSilva and Matrka2
The risk of viral transmission from surgical or percutaneous tracheostomy has led to a reluctance to undertake the procedure, leaving patients intubated for extended time periods. Coupling these airway management factors with a pathology that may increase upper airway inflammation has resulted in an emerging cohort of patients with challenging subglottic and tracheal stenoses, requiring multiple interventions. We present three patients from our tertiary airway centre and discuss future clinical implications.
Medline and PubMed databases were searched for articles highlighting an association between Covid-19 and subglottic or tracheal stenosis. Search terms used included ‘COVID-19’, ‘subglottic stenosis’, ‘tracheal stenosis’ and ‘laryngotracheal stenosis’. One article from Boston, USA, described laryngeal complications in 20 patients recovering with Covid-19.Reference Naunheim, Zhou, Puka, Franco, Carroll and Teng3 The patients were intubated for an average of 21.8 days. Two patients (10 per cent) had a diagnosis of subglottic stenosis.
Case reports
The patients’ demographic details and clinical course are detailed in Table 1. All three patients had a size 8 (inner diameter of 8.0 mm) endotracheal tube and at no point were their tubes downsized. They were intubated for a mean of 28 days (range, 27–48 days) (Table 1). The patients’ intensive care stays and surgical tracheostomy insertions took place at different hospitals, and patients were subsequently referred to our tertiary airway unit. All patients underwent biopsies, which demonstrated inflammation and fibrosis.
Patients’ characteristics and interventions
DM = diabetes mellitus; BMI = body mass index; ETT = endotracheal tube
Case one
Patient one, a 58-year-old male, was intubated for 27 days before surgical tracheostomy. After decannulation, he developed hospital-acquired pneumonia and type II respiratory failure. Re-intubation failed and he needed emergency percutaneous tracheostomy placement. He was again decannulated, and subsequently required intubation because of stridor. Contrast-enhanced computed tomography (CT) of the larynx revealed short-segment stenosis at the level of the third and fourth cartilage (Figure 1). Microlaryngoscopy and bronchoscopy demonstrated tracheal A-frame deformity and further 70 per cent stenosis 20 mm distal to this (Figure 2). The patient was managed with a 40 mg/ml Depo-Medrone® (methylprednisolone acetate) injection, laser cruciate incisions and balloon dilatation to the stenotic segment. He required three further microlaryngoscopy and bronchoscopy, laser, and balloon dilatation procedures, with a mean of 23 days between each procedure. He then underwent an elective tracheal resection of a 20 mm stenotic segment with end-to-end anastomosis.
Axial and sagittal (inset), contrast-enhanced computed tomography images of the larynx in patient one, demonstrating short-segment stenosis at the level of the third and fourth tracheal rings. L = left; P = posterior
Intra-operative image of patient one, who had 70 per cent tracheal stenosis.
Case two
Patient two, a 47-year-old female, was referred to our unit because of failed tracheostomy decannulation. Initial CT revealed no significant subglottic stenosis. Microlaryngoscopy and bronchoscopy demonstrated obstructive sleep apnoea (OSA) caused by oropharyngeal collapsibility, significant laryngopharyngeal reflux related oedema and 30 per cent suprastomal granulation tissue. The granulation tissue was injected with Depo-Medrone 40 mg/ml and the patient's tracheostomy tube was downsized to size 6 (inner diameter of 6.0 mm), but decannulation failed. The patient underwent a repeat microlaryngoscopy and bronchoscopy procedure 19 days later, which demonstrated an anterior subglottic stenosis of 10 mm below the vocal folds. Carbon dioxide (CO2) laser cruciate incisions and balloon dilatation were performed to the stenosis, followed by successful decannulation 5 days later.
Case three
Patient three, a 42-year-old male, had been intubated for 27 days and had had a tracheostomy for 13 days. He presented to the emergency department 8 days post-discharge from hospital with stridor. Computed tomography of the neck revealed tracheal stenosis at the level of the second and third tracheal cartilage. Initial microlaryngoscopy and bronchoscopy demonstrated A-frame deformity and 20 per cent tracheal stenosis, located 45 mm below the vocal folds. The stenosis was treated with intralesional Depo-Medrone injections, CO2 laser cruciate incisions and balloon dilatation up to 18 mm, leaving a normal diameter airway. He presented 30 days later with 30 per cent stenosis and the procedure was repeated. He required two further microlaryngoscopy and bronchoscopy, laser, and balloon dilatation procedures, and is now scheduled for tracheal resection and end-to-end anastomosis.
Discussion
The clinical course of patients with Covid-19 requiring intubation represents a significant challenge in the management of subsequent subglottic and tracheal stenosis. Our case series shows that these patients can present with problems at multiple levels in the subglottis and trachea, requiring multiple interventions for recurrent stenosis. Two patients required tracheal resection and end-to-end anastomosis because of the severity of stenosis and the short time between interventions. Post-tracheostomy A-frame deformity usually responds to endoscopic tracheoplasty,Reference Nouraei, Kapoor, Nouraei, Ghufoor, Howard and Sandhu4 but this treatment does not appear to be as responsive in Covid-19 patients. The mean time between interventions in our case series was 33.9 days, which is a tenth of the mean time between interventions for other causes of subglottic stenosis.Reference Feinstein, Goel, Raghavan, Long, Chhetri and Berke5
Multiple factors contributed to the clinical course of these three Covid-19 patients. One factor is the mean number of days the patients were intubated, which was 28 days (range, 27–48 days). This is a week longer than in a case series from the USA, of 20 Covid-19 patients with laryngeal complications, who were intubated for an average of 21.8 days.Reference Naunheim, Zhou, Puka, Franco, Carroll and Teng3 Timing of tracheostomy for respiratory weaning in the intensive care unit is generally advised to be 7–15 days,Reference Vargas, Sutherasan, Antonelli, Brunetti, Corcione and Laffey6 to avoid long-term complications of intubation and ventilation such as subglottic stenosis. The mean time to tracheostomy in the three patients in our case series was almost double the upper limit of this. Patients with Covid-19 remain intubated for longer, with prolonged time to tracheostomy, because of the risk of aerosol generation during tracheostomy placement and aftercare. This leads to prolonged irritation of the subglottic and tracheal mucosa by endotracheal tubes. Being nursed in a prone position has also been hypothesised to be a risk factor causing laryngeal complications.Reference Naunheim, Zhou, Puka, Franco, Carroll and Teng3
All three patients had large endotracheal and tracheostomy tubes placed (Table 1). Anecdotally, cuff pressures have been kept high in Covid-19 intensive care patients, to reduce risk of aerosolisation from cuff leak.
Co-morbidities also play a role – two patients had diabetes and a high body mass index (BMI), which are risk factors for subglottic stenosis.Reference Nicolli, Carey, Farquhar, Haft, Alfonso and Mirza7 Laryngopharyngeal reflux occurs in obese and intubated patients, and the resultant chronic inflammation is associated with subglottic stenosis.Reference Nicolli, Carey, Farquhar, Haft, Alfonso and Mirza7 High BMI will also compound airway problems associated with OSA and laryngo-tracheomalacia.
We propose that intensive care units standardise the following factors in Covid-19 patients, as in other intensive treatment unit patients: duration of intubation before tracheostomy, endotracheal tube size, and cuff pressures. Otherwise, a large cohort of patients with complex stenoses will emerge that requires multiple high-risk interventions.
• Patients requiring intubation for coronavirus disease 2019 (Covid-19) remain intubated for prolonged periods before tracheostomy
• Critical care patients with Covid-19 often have co-morbidities associated with laryngotracheal stenosis
• Intubated Covid-19 patients are therefore at increased risk of developing multi-level, severe stenoses
• Our cohort of patients indicates that these stenoses are less responsive to endoscopic management
Prolonged intubation, increased time to tracheostomy, high cuff pressures and large endotracheal tubes put patients with Covid-19 at increased risk of subglottic stenosis. The risk is enhanced by upper respiratory inflammation caused by the virus and the co-morbidities associated with Covid-19. Intensive care and otolaryngology teams should collaborate to reduce the risk factors for stenosis in this new cohort of patients.
Competing interests
None declared
