The use of three-dimensional printing and virtual reality to develop a personalised airway plan in a 7.5-year-old child

Editor, Computer-assisted design programmes make anatomically correct models from a patient's computed tomography (CT) images. This model can be printed on a three-dimensional printer or turned into a virtual reality programme used with commercially available headsets. To date, the use of three-dimensional printing in anaesthesia has been mainly for educational purposes.1 Lung isolation for one-lung ventilation (OLV) can be challenging. In paediatric patients, the range of equipment available is limited. The smallest double-lumen tube (DLT) commercially available is 26 French gauge (FG), with an external diameter of 8.6 mm.2 Although it is not generally recommended in children younger than 8 years of age, there are case reports in younger children.3 Despite extensive adult experience, our familiarity with OLV in the very young paediatric population is limited. We were presented with a 7.5-year-old 18-kg child scheduled for right upper lobectomy due to lung metastasis from a Ewing's sarcoma. We used a combination of three-dimensional printing and virtual reality bronchoscopy to develop a personalised airway plan reducing the potential for trial and error in airway manipulation. The patient's family consented to the publication of this report. The three-dimensional model was developed by identifying the trachea, main bronchi and lobar bronchi from an existing CT scan (IntelliSpace Portal v11; Philips Healthcare, Best, The Netherlands). The images were imported into postprocessing software (3-matic medical 14.0; Materialise, Leuven, Belgium). The walls of the model built outwards to maintain the internal diameter and hollowed out so that we could test OLV. The model was then printed in clear plastic (PolyJet J750; Srtatasys, Rehovot, Israel), and converted into a virtual reality program (D2P; 3D Systems, Rock Hill, California, USA; Vive; HTC, Taoyuan, Taiwan; SteamVR; Valve Corporation, Bellevue, Washington, USA) to allow a virtual bronchoscopy to be performed. The CT measurements were reassuring for use of a 26-FG DLT (tracheal diameter 9 mm) but not for a bronchial blocker [distance from the right upper lobe bronchus (RULB) to the carina 9.4 mm] (Fig. 1a). Prior to using the model our airway plan was (1) A: 26-FG DLT (Teleflex, Morrisville, North Carolina, USA); (2) B: 5-mm microlaryngeal tube (MLT) (Smiths Medical, Dublin, Ohio, USA). Fig. 1: Airway planning. (a) Computed tomography measurements. (b) Using the virtual bronchoscopy. (c) The three-dimensional printed airway model. (d) 26-FG double-lumen tube. (e) Size 5.0-mm endotracheal tube directed down the left main bronchus. (f) A 5-FG bronchial blocker. DLT; double-lumen tube; ETT, endotracheal tube; FG, French gauge.The day before surgery, the treating anaesthesiologist (VV) spent time on the virtual reality simulator familiarising himself with the patient's airway anatomy (Fig. 1b). Thereafter, the model was used to formulate an individualised airway plan (Fig. 1c). Immediately, it became obvious that the 26-FG DLT was too big (Fig. 1d). Next, Plan B was tested by inserting the 5-mm MLT into the left main bronchus (LMB) using a 2.8-mm bronchoscope (Karl Stortz, Berlin, Germany) to guide the MLT at the carina. This fitted easily as did a size 5.5-mm MLT (Fig. 1e). We then tried a 5-FG bronchial blocker (Fuji Systems, Tokyo, Japan) through a normal 5-mm cuffed endotracheal tube (ETT) (Smiths Medical, Dublin, Ohio, USA). This technique was not originally part of our plan due to concerns about the balloon obstructing the RULB. On the model, however, this was not the case (Fig. 1f). Finally, the lobes were simulated by attaching fingers of disposable gloves to the lobar bronchi. By ventilating with a reusable resuscitator (Ambu, Ballerup, Denmark), we were able to asses our airway plan for adequate OLV. Both the 5-mm MLT and the bronchial blocker provided adequate OLV but the cuff of the 5.5-mm MLT remained in the trachea. An MLT is 31 cm and the ETT is 21 cm. The predicted insertion depth was 15 cm so there were concerns about the stability of an MLB with bronchial blocker in this patient. As there were no indications for difficult intubation and we were already familiar with use of a bronchial blocker through a standard ETT, our final airway plan was (1) A: 5.0-mm ETT and bronchial blocker (2) B: 5.0-mm MLT directed into the LMB Following induction of anaesthesia, the trachea was intubated with a 5-mm ETT. Lung isolation was achieved on the first attempt using a bronchial blocker with fibreoptic guidance. During the procedure, the treating anaesthesiologist reported a good correlation between the patient's anatomy and the virtual reality model, allowing easy identification of the carina and RULB. The patient was placed in the left lateral position and OLV was commenced. There was complete airway blockade and collapse of the right lung throughout the surgery with no repositioning of the bronchial blocker required. At the end of the procedure, the patient was extubated and the patient made an uneventful recovery (https://links.lww.com/EJA/A283). Three-dimensional printing in anaesthesia is underutilised for patient-specific pre-operative planning.1,4,5 The same applies to the use of virtual reality in anaesthesia.6,7 We successfully combined both technologies to produce a personalised airway plan for a paediatric patient. The final airway plan was substantially different to the plan developed using standard imaging techniques. This reduced the number of attempts at lung isolation that would have been performed on the patient. Virtual reality and three-dimensional printing have been combined before in treatment planning by surgeons and oncologists as part of a multidisciplinary team.8 Despite a good correlation between CT-generated virtual reality bronchoscopy images and actual bronchoscopy findings, this has not been a technology embraced by anaesthetists.9,10 In the current case, the treating anaesthesiologist reported the virtual reality bronchoscopy to be helpful in identifying airway structures and ensuring correct placement of the ETT. In addition, it helped to confirm the position of the RULB when placing the bronchial blocker. In retrospect, it was potentially naïve of us to think that a 26-FG DLT was appropriate based solely on two-dimensional CT measurements. Three-dimensional CT measurements have shown to vary from the two-dimensional measurements. This, combined with the external diameter of the 26-FG DLT not taking into account the tracheal cuff, explains the mismatch we found on the model.11 We deliberately decided to print our model in hard plastic as an added safety factor. We were concerned that if we used a more elastic material, we might use an ETT or DLT that was too big, causing trauma to the patient. We believe this to be one of the few cases reported in which three-dimensional printing has been used for airway planning and the first in which virtual reality has been combined with three-dimensional printing to enhance understanding of patient-specific anatomy.12–15 Teams experienced in paediatric thoracic surgery might find this 'overkill'. For thoracic anaesthesiologists less familiar with OLV in paediatric patients, preparing a personalised airway model can quickly and safely provide additional experience.