Use of three-dimensional printing to guide fibreoptic intubation in a patient with a difficult airway caused by massive adenoid cystic carcinoma of the maxillary bone: A case report

We would like to describe our use of three-dimensional (3D) printing technology to guide the airway management of a patient with a known difficult airway caused by a massive adenoid cystic carcinoma (ACC). Consent for publication was obtained from both the patient and Qingdao Municipal Hospital. The patient also consented to the publication of all images gathered.

A 63-year-old male patient presented to our hospital for treatment of a massive ACC of the right maxillary bone by radioactive particle implantation. He had no more than one finger-breadth mouth opening due to the tumour. Computed tomography (CT) and magnetic resonance imaging (MRI) showed a hyperdense mucosal mass, with a large number of endovascular tumour plugs. The size of the tumour was approximately 10.7 × 9.8 × 9.8 cm, extending from his lips to the base of his tongue and causing severe oropharyngeal narrowing (Figure 1). We initially planned to use a fibreoptic bronchoscope (FOB; 6.0 mm external diameter and 2.8 mm working channel; Olympus, Tokyo, Japan) according to the Difficult Airway Society intubation guidelines. ¹ We titrated the sedation to a level 2 on the Ramsay sedation scale to suppress the intubation reflex. Then, the FOB was primed with a reinforced endotracheal tube (Fuzhou Kanglite Surgical Plastic Cement Co. Ltd., Fuzhou, China) with an internal diameter of 7.0 mm. Despite changing the position of the reinforced endotracheal tube continuously, manipulation or rotation could not improve the view sufficiently for us to identify any laryngeal structure. As the FOB was advanced again, the patient immediately began secreting mucus from his oral cavity, which completely blocked the view despite continuous suctioning and injecting of atropine (0.5 mg) in advance. Unfortunately, we were unable to see any airway structures due to the secretion of mucus. The decision was made to proceed with the lightwand (Jerome, Shanghai, China). The lightwand was introduced into the mouth, pushed, rotated and moved, until a bright, well-defined aperture appeared in the neck. A 7.0 mm (internal diameter) reinforced endotracheal tube was then successfully inserted over five minutes using the lightwand.

Figure 1.

Computed tomography (CT) and magnetic resonance imaging (MRI) manifestations of the tumour. (a) A head CT scan showed the size of the tumour was approximately 10.7 × 9.8 × 9.8 cm (b) A head MRI scan showed the condition of the tumour and surrounding tissues.

As the patient required subsequent treatments requiring intubation, we felt that it was important to assess the stenosis of the patient’s respiratory tract in more detail, and decided that it would be helpful to simulate the fibreoptic intubation exercise in advance. We therefore constructed a 3D model of the patient’s airway using a 3D printer (Stratasys, Eden Prairie, MN, USA) to allow for detailed assessment and practice.

Data from 64-slice spiral CT scans (GE Medical Systems, Bloomington, IL, USA) of the head were saved as digital imaging and communications in medicine (DICOM) format, which were evaluated by author LS to identify respiratory anatomical features and areas of the tumour. Raw images were put into Materialise Mimics Innovation Suite 22.0 research (Leuven, Belgium) to reconstruct 3D images of the area of the head and neck, especially the respiratory tract according to a threshold-based algorithm (Figure 2). The 3D images were exported to assess and analyse in a panning or rotation way the shape of the respiratory tract and the degree of respiratory tract compression. The imaging data, saved in stereo lithography format, was shaped and refined into a 1:1 ratio respiratory tract prototype by a 3D printer (Stratasys, Eden Prairie, MN, USA) (Figure 3). The 3D models of the respiratory tract site were manufactured in just 14 hours. With precise data for the diameter of the narrow upper airway and repeated fibreoptic intubation exercises on the 3D model, we discovered that it would accommodate a 7.0 mm (internal diameter) reinforced endotracheal tube.

Figure 2.

The preoperative situation. (a) The adenoid cystic carcinoma (ACC) in the mental region occupied most of the floor of the mouth as seen from the front. (b) Tridimensional images rendered model of the head and neck. (c) Tridimensional printed model of the head and neck.

Figure 3.

The printed model of the respiratory tract. (a) Tridimensional images rendered model of the head and neck including the respiratory tract. (b) Tridimensional images rendered model of the respiratory tract. (c) Tridimensional printed model of the respiratory tract as seen from the front. (d) Tridimensional printed model of the respiratory tract as seen from the back.

For the second procedure, the patient was fasted as normal, and the tongue and posterior pharyngeal wall were anaesthetised with topical lidocaine. In addition, 4 ml of 2% lidocaine was injected into the larynx via a cricothyroid membrane puncture. Sedation was provided with dexmedetomidine (1 μg/kg intravenously (i.v.) over 10 minutes) followed by an infusion of dexmedetomidine (1 μg/kg/hour). This was supplemented with sufentanil 5 μg i.v. to suppress the intubation reflex. This local anaesthesia and sedation were the same as for the first procedure. The fibreoptic intubation directed by 3D models was much easier to achieve than in the first operation thanks to the intubation simulation training. We saw the carina in only three minutes and the stimulation of the airway mucous membrane was reduced because the intubation time was greatly shortened. The patient was maintaining 100% oxygen saturation and was not in any obvious discomfort. The tube was connected to the ventilator, and its placement was evaluated by auscultation of bilateral breath sounds and capnography. Subsequently, general anaesthesia was maintained with propofol, rocuronium and remifentanil infusion. Surgery and extubation while fully awake were uneventful. We also successfully intubated the patient with the same technique under sedation for his recent third procedure, and the operation went smoothly.

In this case, we found the 3D model to be helpful in assessing the anatomy of the patient’s airway, to decide what size tracheal tube to use, and to practise the fibreoptic intubation procedure with tactile feedback. Moreover, the 3D model enabled better communication with the patient and his family about intubation plans. We feel that the 3D model improved the airway assessment (versus. relying on two-dimensional CT and MRI images alone), saved subsequent operation time, and potentially reduced bleeding caused by tissue damage during the intubation process. It is necessary to point out that the 3D casted airway models are manufactured by the placement of computer-aided design materials layer by layer, so it is economical and practical to apply in predictable difficult airways.

In recent years, 3D imaging and printing technology, known as rapid prototyping, has become a new clinical application, which is widely used in preoperation evaluation or for surgical teaching,^2–4 but its application for difficult airways is sparse. From this case, we realised that tracheal intubation simulation training with 3D printing for the management of a predictable difficult airway may assist us in saving intubation time and improving the overall success rate. The emergence of the 3D printing technique makes predictable difficult airways ‘visible’.