Tracheal tube obstruction as a complication of transoesophageal echocardiography

Summary

Complications of transoesophageal echocardiography are numerous and may have serious consequences. We present the case of a 31‐year‐old woman with postoperative airway obstruction secondary to a transesophageal echocardiography probe. The patient had been admitted with acute myocarditis and required temporary mechanical support with a biventricular assist device. She deteriorated on the intensive care unit several hours later with hypoxaemia, high airway pressures and reduced tidal volumes. Sedation was adequate and no external obstruction in the breathing circuit was observed. The tracheal tube was noted to be permanently deformed in the oropharynx, causing airway obstruction. Tracheal tube exchange was required, and the patient recovered from the event. We suspect that the position of the transoesophageal echocardiography probe in the operating theatre had contributed to the deformity, and the presence of airway obstruction was masked by the reduced ventilatory parameters instituted while on mechanical circulatory support. The biventricular assistance devices were explanted subsequently, and the patient was discharged home on day 31. This is the first reported case of a kinked tracheal tube where transoesophageal echocardiography probe placement was suspected to have contributed. A high index of suspicion is required for this complication on the intensive care unit.

Introduction

Transoesophageal echocardiography (TOE) is a standard peri‐operative monitoring tool used in cardiac surgery. Recent data suggest that complications are more common than previously thought, with significant morbidity. Recognition and diagnosis may be difficult [1]. Potential complications of TOE are numerous and can be life threatening [2]. Airway complications alone are considerable.

We present a case of significant tracheal tube obstruction secondary to deformation and kinking. It is suspected that intra‐operative TOE contributed to this phenomenon.

Report

A healthy 31‐year‐old woman (height 162 cm; weight 66 kg; BMI 25.1 kg.m^‐2) presented to hospital with chest pain and dyspnoea in July 2020. There was severe left ventricular systolic dysfunction seen on transthoracic echocardiography. Coronary angiography was normal. Right heart catheterisation and biopsies were performed following transfer to a tertiary centre. The patient received mechanical support via an intra‐aortic balloon pump, followed by urgent biventricular assist device (BiVAD) implantation the following day. An oxygenator was included in the left‐sided circuit for additional support. The patient returned to the cardiothoracic ICU from the operating theatre, with her trachea intubated and her lungs ventilated on rest settings, in the early hours of the next morning. A provisional diagnosis of viral cardiomyopathy was made and suitability for transplant was assessed.

The patient underwent a trial off the circuit oxygenator that afternoon. Adrenaline, milrinone and dopamine infusions had been weaned, with remaining circulatory support consisting of noradrenaline (0.12 µg.kg^‐1.min^‐1) and vasopressin (0.015 u.kg^‐1.h^‐1). Over the following hours, the ventilatory requirement remained minimal and F_IO₂ was weaned to 0.3. The patient was deeply sedated with a Richmond Agitation Sedation Score of −4 achieved with intravenous (i.v.) propofol and alfentanil infusions.

Two hours later, the nursing staff were unable to pass a suction catheter through the tracheal tube. Peripheral oxygen saturations dropped to 92% and tidal volumes reduced to 300 ml (Table 1). There had been no premonitory signs of impaired ventilation. The patient's heart rate was 114 beats per min and temperature 36.3°C. Increased sedation, a bite guard and repositioning did not improve ventilation. There was no apparent patient effort to bite the tracheal tube and no visible circuit occlusion. There was firm obstruction to attempted closed‐circuit suctioning of the airway approximately 10 cm from the circuit end. The tracheal tube was felt to be deformed and kinked at an acute angle at the oropharynx. The team made a balanced decision for re‐intubation to relieve airway obstruction rather than providing ongoing increased support through escalation of positive pressure. The oxygenator was restarted alongside conventional pre‐oxygenation through the deformed tube, given that exchange over a bougie would not have been possible. The patient’s trachea was extubated and re‐intubated uneventfully using a videolaryngoscope and bougie. The original tracheal tube was found to be permanently deformed at 15 cm from the tip, consistent with its position in the oropharynx (Fig. 1).

Table 1.

Changes in postoperative respiratory settings and observations.

Event number	1	2	3	4	5
Time;h	0	11	17	20	23
Settings
FIO2 (ventilator)	0.6	0.4	0.3	0.4	0.3
PEEP; cmH2O	5	5	5	5	5
Pinsp; cmH2O	15	15	15	15	15
Sweep; l.min‐1	2	‐	‐	‐	‐
FIO2 (BiVAD)	0.65	‐	‐	‐	‐
Observations
SpO2; %	100	100	98	92	97
PaO2; kPa	19.6	23.4	14.6	12.5	19.6
VT; ml	371	367	376	300	376
VT; ml.kg‐1	6.8	6.8	6.9	5.5	6.9
EIP; cmH2O	17	19	21	34	20
RR; breaths.min‐1	20	15	14	18	20
Suction possible	Yes	Yes	Yes	No	Yes

Events: 1, return from the operating theatre; 2, trial off biventricular assist device oxygenator; 3, relative stability; 4, observation of decreased tidal volume; 5, stability following tracheal tube change.

PEEP, positive end‐expiratory pressure; P_insp, inspiratory pressure; BiVAD, biventricular assist device; V_T, tidal volume (ideal body weight 54.2 kg used); EIP, end‐inspiratory pressure; RR, respiratory rate.

Figure 1.

The tracheal tube showing deformity after removal at approximately 15 cm. It was a polyvinyl chloride Portex® (Smiths Medical, Minneapolis, USA) tracheal tube with an internal diameter of 8.0 mm and external diameter 10.9 mm. The cuff inflation tube exits more distally at approximately 20 cm.

The patient’s condition improved quickly, and ventilation was reduced to previous requirements. Other causes of obstruction were not identified on chest radiography. It was suspected that the tracheal tube had been deformed by the TOE probe in the operating theatre, with a delayed presentation several hours later in the ICU as the kink evolved further. The BiVAD was explanted on postoperative day 24 after setbacks with pericardial effusion and pneumonia, before the patient was discharged from hospital on day 31. She was later seen at the clinic with improved left ventricular systolic function. The diagnosis was lymphocytic myocarditis.

Discussion

This is the first case to be described in the literature in which permanent kinking of the tracheal tube and subsequent airway obstruction is suspected to have been contributed to by a TOE probe. Kinking of the tracheal tube by the probe has been well documented in paediatric practice, where thin‐walled tubes are commonplace [5]. Reports exist in the adult population, although these appear only to describe native airway occlusion where probe flexion against the lower airways has been implicated, with or without anatomical variations [3, 4].

A review of tracheal tube incidents illustrates a variety of aetiologies of tube kinking: tube holder attachment; intra‐oral kinking; patient biting; Boyle‐Davis gag; and nasogastric tube insertion [6]. Intra‐oral kinking around the 18–20‐cm markings may be due to localised weakness at the exit point of the cuff inflation tube or significant tracheal tube flexion [7]. One report described transient intra‐oral kinking which disappeared after extubation [8]. Another report of intra‐operative kinking at 10 cm from the tip was attributed to the combination of thermal effects with malpositioning of the tracheal tube against its natural curvature [9].

The mechanism of deformity in this case is postulated to have involved a combination of heating by the patient and the mechanical effects of TOE probe position. Most tracheal tubes are made of polyvinyl chloride, as in this case. Thermal softening can occur at as low as 36°C secondary to local tissue heat or exogenous sources, such as active patient warming systems [7, 10]. Probes are manufactured to comply with international standard IEC60601‐2‐37, in which their temperature (notably at the tip) should not exceed 43°C. The area of deformity at 15 cm in this case is in keeping with the position of the oropharynx, where the shaft of the TOE probe was adjacent to the tracheal tube for several hours. The position of the probe is likely to have facilitated kinking at this point.

There was a period of compensation before sudden deterioration in this case. The presence of the oxygenator in the BiVAD circuit and lung rest may have contributed to recognition being delayed. We suggest that there was some element of softening or weakening of the tube adjacent to the TOE probe in the operating theatre. This weak point in the tracheal tube might have been exacerbated by thermal softening on the ICU. Existing reports demonstrate kinking can occur at a range of times from intubation and can be caused by a variety of factors, but data are limited [8, 9, 10]. The involvement of the TOE probe is supported by the unique position at which the kink occurred, away from other causes of weakness such as the pilot tube exit point.

This case demonstrates a potentially serious cause of airway obstruction in the ICU. It is notable that the presentation was several hours after the patient’s return from the operating theatre and a high index of suspicion may be required for postoperative complications of transoesophageal echocardiography.