Unexpected difficult tracheal intubation secondary to subglottic stenosis leading to emergency front‐of‐neck airway

Summary

A 26‐year‐old woman presenting for an elective day case procedure under general anaesthesia had undiagnosed subglottic stenosis leading to a life threatening airway emergency requiring emergency front‐of‐neck airway. We outline the case and discuss key anaesthetic considerations in subglottic stenosis, including concerning features of a medical/anaesthetic history and the potential for rapid deterioration of a stenotic airway following manipulation. We also consider the effect of anaesthesia on the calibre of subglottic stenosis and the effects of positive pressure ventilation. Subglottic stenosis is a rare condition with congenital, acquired and idiopathic origins; however, iatrogenic trauma is the most common cause. We are aware of a small number of published case reports of previously undiagnosed subglottic stenosis in adults discovered after induction of anaesthesia; situational deterioration to ‘cannot intubate, cannot oxygenate’ scenarios appear even rarer.

Introduction

The subglottic space extends from the inferior margin of the vocal cords to the lower border of the cricoid cartilage. Subglottic stenosis is a narrowing of the trachea between these anatomical landmarks and although it can be due to a variety of causes, with injury following tracheal intubation or tracheostomy being the most common causes. It may result from a prolonged period of tracheal intubation in the ICU [1].

Mild subglottic stenosis is often asymptomatic; however, when present the main reported symptoms are dyspnoea (especially after exertion), hoarseness and occasionally stridor. Subglottic stenosis is frequently misdiagnosed as asthma and patients are often referred to ear, nose and throat (ENT) surgeons only once asthma medications have failed to alleviate their symptoms [2].

A focused literature review revealed only six case reports of previously unrecognised subglottic stenosis in adults discovered following induction of anaesthesia. None describe failed tracheal intubation leading to emergency front‐of ‐neck airway (eFONA). The reports describe a range of origins of subglottic stenosis including post‐intubation injury, Wegener’s granulomatosis, cricoid chondrosarcoma, laryngeal web, subglottic ring and subglottic stenosis related to pre‐eclampsia in pregnancy.

These case reports identify several themes which are common to this case. These include the poor predictive value of conventional difficult airway screening methods, a history of mild dyspnoea which had been attributed to asthma and the likelihood that the anaesthetist will need to employ a variety of techniques in order to safely manage the airway.

Report

A 26‐year‐old woman, with depression and a body mass index of 32 kg.m^‐2, attended for a day case hysteroscopy and endometrial biopsy under general anaesthesia. Prescribed medications included a beclometasone dipropionate/formoterol fumarate dihydrate inhaler and fluoxetine. She had undergone an uncomplicated laparoscopic cholecystectomy 8 years prior.

Only volume two of her notes were available to the list anaesthetist. Airway examination was unremarkable. Medical history revealed exertional dyspnoea, specifically after climbing two flights of stairs. Symptoms were attributed to asthma, which the patient had felt were unresponsive to β₂ agonists. Consequently, the patient had discontinued her inhalers.

Following induction of anaesthesia with propofol and fentanyl, a supraglottic airway device (size 4 i‐gel^®, Intersurgical, Wokingham, UK) was inserted and manual ventilation deemed possible. Bilateral air entry was confirmed on auscultation with normal end‐tidal capnography. Anaesthesia was maintained with sevoflurane. She was transferred to operating theatre and her lungs ventilated using synchronised intermitted mechanical ventilation while apnoeic. Initial positive pressure ventilation resulted in high peak inspiratory pressures (40 cmH₂O) with low tidal volumes (120 ml). Further propofol was administered and the i‐gel exchanged for a size 5, with no improvement.

The patient was paralysed for tracheal intubation. Direct laryngoscopy revealed a grade‐1 view of the glottis; however, only the tip of a size 7.0 oral tracheal tube could be advanced through the vocal cords. On second attempt a gum elastic bougie was passed without difficulty into the trachea, but again, the tracheal tube could not be advanced. Two consultant anaesthetists made three further unsuccessful attempts at tracheal intubation: first, with a size 7.0 tracheal tube and optimised positioning; second with a size 5.5 tracheal tube; and third with a flexible bronchoscope (Ascope™ 4 Broncho Slim, Ambu A/S, Ballerup, Denmark) which could not be advanced beyond the vocal cords and demonstrated extensive subglottic oedema.

Oxygenation was maintained with bag‐mask ventilation which became increasingly difficult, eventually requiring a four‐handed technique with increasing end‐tidal carbon dioxide (> 10 kPa). A decision was made to attempt an eFONA. Cricothyroidotomy and subsequent tracheostomy were successfully undertaken.

The patient was transferred to ICU. Diagnostic microlaryngoscopy 1 month later found a grossly oedematous supraglottis with extensive granulation tissue of the right vocal cord, false cord and immediate subglottis. A biopsy showed mild, non‐specific chronic inflammation. One year later, the patient has undergone multiple treatments for Cotton grade 3 tracheal stenosis using balloon dilatation and carbon dioxide laser, but due to consistent restenosis, it has not been possible to decannulate her (Fig. 1).

Figure 1.

Microlaryngoscopy images from different timepoints after the event showing (a) diagnostic microlaryngoscopy (1 month following event); (b) repeat microlaryngoscopy due to difficulty weaning showing a stricture at the level of the vocal cords (5 months following the event); (c) at time of balloon dilation showing ongoing stenosis (7 months following the event); and (d) following several balloon dilations and carbon dioxide laser showing the subglottic region to be more open with persistent glottic stenosis (15 months following the event).

It was subsequently discovered that at the age of 19 the patient had an ICU admission for a polypharmacy overdose requiring a tracheostomy. Details of this admission were in volume one of her notes. During this admission, she had four failed tracheal extubation attempts. Rigid bronchoscopy 2 days before this initial tracheostomy documented no evidence of subglottic stenosis. The tracheostomy was removed 8 days later without complication. She was not followed‐up after her ICU discharge.

On retrospective questioning, the patient noticed dyspnoea immediately following discharge from ICU which had been deteriorating before this admission. She had seen her general practitioner on multiple occasions and was diagnosed with asthma which proved unresponsive to β₂ agonists. No association between her dyspnoea and her ICU admission had been identified. The patient reported having sought ENT review actively, including privately at the time of her hysteroscopy.

Discussion

Subglottic stenosis is a rare condition that can be congenital, acquired or idiopathic in origin. Post‐intubation injury by tracheal tube or tracheostomy are the most common causes. This patient likely developed subglottic stenosis following an extended ICU stay where she failed several tracheal extubation attempts over an 18‐day period and subsequently required a tracheostomy.

The Fourth National Audit Project reports two cases of unanticipated difficult tracheal intubation due to undiagnosed tracheal stenosis, one resulting in urgent surgical tracheostomy [3].

Patients with acquired subglottic stenosis are normally diagnosed from a few days to 10 years or more following the initial injury [1]. Two case reports describe the development of severe, acute subglottic stenosis in young adults (aged < 30 y) following a period of tracheal intubation within the preceding 6 months [4, 5]. However, more commonly subglottic stenosis is not diagnosed for a considerable period of time after the initial injury as patients are either asymptomatic or only mildly dyspnoeic on exertion. As was the case with this patient, subglottic stenosis is often misdiagnosed as asthma and treated with bronchodilators [6, 7, 8], therefore medical history can be a poor predictor of subglottic stenosis, as can conventional airway assessment.

A high index of suspicion is needed with regard to patients who have spent time with their tracheas intubated or with a tracheostomy in situ, and subsequently describe new or worsening respiratory symptoms. Points on clinical history and examination that may heighten suspicion of subglottic stenosis are set out in Table 1. A thorough pre‐operative history and any time spent in ICU (including medical admissions) is important in identifying these patients – and a potential source of oversight in young, fit individuals. A tracheostomy scar was missed in the case we report which, if noticed, may have prompted further questioning.

Table 1.

Points on history and examination that may raise suspicion of undiagnosed subglottic stenosis.

History
Asthma unresponsive to β2 agonists
Previous ICU admission with difficult tracheal intubation, failed tracheal extubation and/or tracheostomy with new or worsening respiratory/ENT symptoms, which may include: Exertional dyspnoea – unresponsive/poor response to β2 agonist treatment Hoarseness Throat discomfort
Examination Tracheostomy scar Stridor

Induction of anaesthesia and neuromuscular blockade leads to loss of intrinsic muscle tone, potentially decreasing the calibre of the stenosis. Negative inspiratory intratracheal pressure from spontaneous ventilation exacerbates the lesion whereas positive pressure ventilation generates positive intratracheal pressure improving ventilation [9]. However, decreasing calibre of stenosis requires increasing pressures to generate flow through the narrowed lumen. This can lead to inadequate ventilation and low tidal volumes due to an inability to generate the required pressure and flow.

Management of unanticipated difficult tracheal intubation should follow guidelines such as those produced by the Difficult Airway Society (DAS), who recommend 3 + 1 attempts at tracheal intubation in order to limit airway interventions and prevent airway trauma [10]. Several case reports support the conclusion that when faced with an unanticipated difficult tracheal intubation but a good view of the glottis, anaesthetists will respond by making more than one attempt at tracheal intubation using smaller diameter tracheal tubes.

A report of a patient with post‐intubation subglottic stenosis undergoing total knee replacement describes two failed tracheal intubation attempts with size 7.0 and 6.5 tracheal tubes, before flexible bronchoscopy revealed a subglottic septum [8]. Supraglottic airway device insertion provided inadequate ventilation and an early decision was made to wake the patient. Another case of a subglottic ring, details several tracheal intubation attempts before successful insertion of a size 5.0 tracheal tube and uncomplicated tracheal extubation [7].

In this case report, four attempts were made at tracheal intubation with one further attempt to view the subglottis with a flexible bronchoscope, by which time the airway obstruction was severe. In retrospect, it is highly probable that multiple attempts at tracheal intubation exacerbated airway obstruction. DAS guidelines state that “when both tracheal intubation and supraglottic airway device insertion have failed, waking the patient is the default option” [10]. However, as demonstrated in this case, progressive airway obstruction in aggravated subglottic stenosis with superimposed airway oedema, may preclude waking the patient.

Of the case reports examined, only one resulted in a situation where it was not possible to wake the patient [6]. This was an obstetric report of unrecognised idiopathic subglottic stenosis which described failure to intubate the patient's trachea with a size 7.0 and 6.0 tracheal tube. Placement of a supraglottic airway device produced evidence of marked airway obstruction. Due to fetal compromise, the delivery proceeded with this airway in situ and the patient regained spontaneous ventilation. Low tidal volumes (100 ml) and high end‐tidal carbon dioxide (12 kPa) meant that waking the patient was unlikely to be successful. Tracheal intubation was finally achieved with a size 4.0 tracheal tube inserted with "considerable force". The patient's trachea was unable to be extubated in ICU and subsequently received a tracheostomy under elective conditions. This case report has the most similarities to the airway deterioration seen in the case we present, with the exception that tracheal intubation was achievable.

The patient we report was not selected for attendance at a formal anaesthetic pre‐assessment clinic due to limitations in information gathered by our screening questionnaire. Contributing external factors included limited notes available on the day of operation with poorly defined volumes, a single anaesthetist assigned to a busy all day list and limited time for pre‐operative assessment. Should a history of severe dyspnoea have been elicited pre‐operatively the patient would have been cancelled and further investigated. This would have included an ENT review and consideration of regional anaesthesia techniques.

Initial failure to advance a tracheal tube through well visualised vocal cords due to resistance should raise suspicion of subglottic stenosis. If faced with this situation, we would advocate waking the patient after a maximum of three attempts at tracheal intubation, including trial with a smaller calibre tracheal tube and adjunct (e.g. bougie). Although the multiple attempts at tracheal intubation are unlikely to have contributed to the long‐term tracheostomy of the patient we present, they likely played a role in the acute deterioration and need for eFONA. The idiosyncratic nature of granulation tissue formation means the presence of a second tracheostomy in this patient likely potentiated an already progressive pathological process.