Prolonged Intubation and Tracheostomy in COVID-19 Survivors

Introduction

COVID-19 was declared a worldwide pandemic in March 2020, with the virus SARS-CoV-2 causing severe acute respiratory syndrome (Williamson et al. 2020). Among patients hospitalised with COVID-19, up to one quarter required Intensive Care Unit (ICU) admission (Huang et al. 2020). For those where COVID-19 led to severe respiratory disease secondary to hypoxaemic respiratory insufficiency or failure, invasive ventilation via endotracheal tube (ETT) was required (Meng et al. 2020). Laryngeal injury following ETT intubation has previously been widely cited as a transient or lasting complication impacting on voice and swallowing and functional outcome (Brodsky et al. 2018; Macht et al. 2011; Schefold et al. 2017; Wallace and McGrath 2020).

Recovery from laryngeal injury and related dysfunction must be managed by a highly skilled multidisciplinary team (Brodsky 2020). Further insult to a fragile respiratory system from laryngeal dysfunction can lead to serious consequences, including aspiration pneumonia, delayed tracheostomy weaning and lack of functional communication. An emerging, distinctive characteristic of the COVID-19 cohort is the duration of ventilator reliance, reported as high as 20 days (Stam et al. 2020). Prolonged intubation is paired with delayed insertion of tracheostomies, increasing the risk of laryngeal trauma. Furthermore, it is possible that the oropharyngeal symptoms of COVID-19, such as cough, loss of taste/smell and pain in the pharynx may have an additional impact on laryngeal function (Lovato et al. 2020; El-Anwar et al. 2020). 

Laryngeal Injury and Dysfunction During Critical Illness 

Intubation Phase

Endotracheal intubation in patients with COVID-19 is a high-risk procedure for staff, requiring full personal protective equipment (PPE) which may impact on the procedure (Cook et al. 2020). Furthermore, airway oedema in those with COVID-19 has been highlighted and anecdotal reports of difficult intubations have also been discussed (McGrath et al. 2020). In these circumstances, laryngeal injury such as mucosal trauma and damage to anatomical structures may be heightened.

The ETT sits in a vulnerable anatomical region for laryngeal function. As the ETT tube is inserted, it passes the arytenoid cartilages, cricoarytenoid joints and vocal folds which are all susceptible to trauma and residual laryngeal complications, manifesting as airway, voice and swallowing impairments (Mota et al. 2012). The recurrent laryngeal nerve innervating the laryngeal musculature is vulnerable to compression by the tube cuff especially if the cuff sits too high, or the cuff pressure exceeds capillary perfusion pressure (Miles et al. 2018). A recent study in over 200 patients intubated for more than 48 hours found that larger tube size was associated with an increased risk of aspiration and laryngeal granulation tissue (Krusciunas et al. 2020).

For those with COVID-19, additional risk factors contributing to laryngeal injury have not yet been investigated. Characteristics that may predispose patients with COVID-19 to neuropathy include age, obesity, diabetes mellitus, hypertension, corticosteroids, extracorporeal membrane oxygenation and laryngopharyngeal reflux (Tadié et al. 2010; Williamson et al. 2020). Ponfick et al. (2015) found pathologic swallowing in 91% and hypoaesthesia of the larynx in 77% of patients with critical illness polyneuropathy. While the sedative management of COVID-19 patients during the intubation phase often precludes verbal communication, during the sedation hold phase some patients may attempt to vocalise while the ETT remains in situ, posing further risk of laryngeal trauma. 

It is well known that prolonged intubation results in laryngeal complications but even transient intubation has been shown to cause mucosal trauma and laryngeal injury (Ng et al. 2019). Prolonged intubation is also known to correlate with post-extubation complications, including laryngeal stenosis, with a reported incidence of 5% in those intubated for 6-10 days (Bonvento et al. 2017). Piazza et.al. (2020) highlights the need for a high-level of suspicion for laryngotracheal stenosis development in COVID-19 patients following long-term intubation and tracheostomy. A description of stenosis in a COVID-19 patient is detailed in the case study below. Laryngeal oedema may negatively impact decision making regarding the safety of extubation, and increase the need for tracheostomy insertion (McGrath et al. 2020). In the COVID-19 RECOVERY trial, critically ill patients assigned to receive dexamethasone corticosteroid resulted in a lower 28-day mortality (Horby et al. 2020). Whilst the use of the dexamethasone may reduce oedema and airway complications, the reported side-effects such as hyperglycaemia, weakness and delirium may further impair communication, laryngeal and swallowing functions.

Prone ventilation has been used as an adjuvant therapy for treatment of acute respiratory distress in those with COVID-19, with some patients being proned repeatedly for up to 16 hours (Zang et al. 2020). Whilst adverse effects such as nerve palsies, pressure ulcers and oropharyngeal swelling are reported, the direct effect of proning on the larynx is not fully understood (Kwee et al. 2015). However, it may be hypothesised that the pressure exerted by the ETT on the laryngeal mucosa may exacerbate laryngeal dysfunction. Furthermore, prolonged bed rest, immobilisation and critical illness are key causes of muscle wasting and loss in ICU patients (Koukorikos et al. 2014; Parry and Puthucheary 2015). Up to 30% muscle mass loss is reported to occur within the first 10 days, however in patients undergoing prone ventilation this may be expedited (Kortebein et al. 2008). A combination of these factors may contribute to sarcopenia-related dysphagia, which has been demonstrated in elderly patients (Zhao et al. 2018). Generalised decline in muscle mass can coincide with weakening of the swallowing musculature and whilst the average age of ICU admissions is lower for COVID-19 patients, the potential impact of muscle wasting should remain a consideration and an age-related red flag. 

Tracheostomy Phase

The timing of tracheostomy insertion in the critically ill COVID-19 patient has been a key area of discussion. For patients with COVID-19, tracheostomy insertion was deemed a high risk and an aerosol generating procedure (AGP) (ENT UK 2020). The balance between delaying insertion to reduce risks to staff during the patient’s most infective period and the extended duration of intubation has been discussed (McGrath et al. 2020). While prolonged ventilation is often necessitated in the COVID-19 cohort, clinicians should remain cognisant of the long term effects of prolonged intubation on laryngeal function. Following insertion, the presence of a tracheostomy tube and inflated cuff can impact upper airway sensitivity, respiratory/swallowing synchronisation and lead to disuse atrophy of laryngopharyngeal musculature (Garuti et al. 2014). Careful assessment of secretion management, cuff deflation tolerance and cough effectiveness by the multidisciplinary tracheostomy team should aim to determine risks to and optimise recovery of pulmonary function in the COVID-19 patient (Garuti et al. 2014). In-depth assessment by SLTs using Fibreoptic Endoscopic Evaluation of Swallowing (FEES) is the most accurate method for assessing oropharyngeal secretions, detecting and managing aspiration risks and for guiding multidisciplinary management and rehabilitation of laryngeal complications (Hafner et al. 2008; McGrath and Wallace 2014; Scheel et al. 2015).

While respiratory compromise associated with COVID-19 has been the most frequently reported, concomitant central and peripheral nervous system impairments have also been detailed (Paterson et al. 2020). Encephalopathy, encephalitis, ischaemic stroke and Guillain-Barré syndrome have all been observed in patients with COVID-19 (Paterson et al. 2020). The presence of central and peripheral nerve involvement further increases the risk of neurogenic laryngeal dysfunction (McIntyre et al. 2020; Schefold et al. 2017). Glossopharyngeal and vagus nerve neuropathies have been identified in a case study of a severely ill COVID-19 patient (Aoyagi et al. 2020). Cranial nerve impairments may exacerbate laryngeal dysfunction and these issues frequently delay recovery times.

Delirium and COVID-19

The inability to vocalise during critical illness can be a significant contributing factor to delirium, anxiety and psychological distress. A range of studies report this as one of the most frustrating and anxious experiences for mechanically ventilated patients (Ford and Martin-Harris 2016). Hospital policies restricting family visitations due to infection control precautions during the pandemic may further contribute to feelings of isolation and disorientation. When unable to vocalise patients have reported feeling trapped, caged, and a loss of personhood and control (Ford and Martin-Harris 2016).

Delirium occurs as a consequence of direct central nervous system invasion as a secondary effect of multiple organ system failure, sedation or environmental factors such as staff wearing PPE (Kotfis et al. 2020). Early restoration of voice can reduce anxiety levels in patients with tracheostomies (Liney et al. 2019). Early evaluation by the SLT and physiotherapist, to assess readiness for cuff deflation or a one way valve application facilitates greater patient participation in treatment. Input from clinical psychology for those experiencing psychological distress has been highlighted previously in the ICU cohort and remains essential in the COVID-19 population (GPICS 2019). Considerations of late sequelae of mental health issues such as anxiety should also be considered by the MDT in the rehabilitation and follow up clinic phases. 

Recovery of Laryngeal Function - ICU Treatment Options

Facilitating Laryngeal Airflow 

For tracheostomised patients, cuff deflation followed by placement of a one way valve provides increased Positive End Expiratory Pressure (PEEP), upper airway sensation and restoration of the ability to verbally communicate. Evaluation of one-way valve tolerance and resultant voice quality by SLTs enables the multidisciplinary team to be aware of any potential laryngeal impairments. Intensivists, nurses, psychologists, dietitians and physiotherapists may all utilise the one-way valve during their interactions with the patient, restoring the opportunity for verbal communication between the clinician and the patient. The use of a one way valve by the patient during video calls to family members who have been unable to remain at the bedside has been highlighted as a significantly positive experience by patients, families and staff alike during the pandemic.