Embracing the robotic revolution into anaesthetic practice

Abstract

‘It is not the strongest species that survive, nor the most intelligent, but the ones most responsive to change’ —Charles Darwin Three years after Christopher Columbus first set sail to the Americas, Leonardo da Vinci designed one of the first humanoid robots in 1495: a mechanical, armoured knight used for entertaining royalty 1. Whilst da Vinci's robot was far from the first example of automated devices (examples are reported from the 1100s 2), mainstream uptake has only recently occurred. Today, there are countless implementations of robotic technology in all guises of life, including the multiple facets of healthcare, but until now there has been little development in automation in the setting of airway management. In this issue of Anaesthesia, Biro et al. have published their proof-of-concept manikin study exploring whether the use of robotic endoscope automated via laryngeal imaging for tracheal intubation (REALITI) can facilitate tracheal intubation 3. This video-endoscope allows manual movement of the endoscopic tip using a joystick at the proximal end of the device, much like a traditional flexible bronchoscope. The device also utilises image-recognition software which can allow 360° circular automated movements of the tip towards the glottis when characteristic anatomical features are recognised. Once the automated mode has been activated, the tip moves in the direction of the geometrical centre point of the glottis. In this study, seven anaesthetists and seven lay persons performed six consecutive device insertions into the trachea of an airway manikin using the joystick, followed by six tracheal intubations using the automatic mode. The key findings were that the automated mode was associated with a faster time to access the trachea compared with the manual mode. Similarly, as users gained more experience with the device, they achieved this desired endpoint faster. The authors report that these findings support the aim of the REALITI project which is to increase successful tracheal intubations by anaesthetists and non-anaesthetists alike. This study suffers several limitations which many readers will easily spot. Firstly, as noted by the authors, tracheal intubation was not actually performed. This is the key endpoint that many studies use to assess the clinical efficacy of devices, both in manikin and patient studies 4, and perhaps should have been the primary metric reported in this study. Indeed, this somewhat undermines the final two words of the device's name. Secondly, the authors compared two arms, both using the REALITI device, but did not compare traditional methods for tracheal intubation with direct laryngoscopy, videolaryngoscopy or flexible bronchoscopy. This would be a logical next step, but its omission in this provisional phase of development is significant. And finally, the comparison of manual and automated modes only demonstrated a speed difference of 5.5 s, which has no real clinical relevance. However, the novelty of this study lies in the ability of the REALITI device to recognise glottic features and subsequently steer the endoscope tip into the trachea automatically. This is the first application of such technology in anaesthetic practice and despite clinical practice still being far off from robot-assisted or automated tracheal intubation, it is a step closer to that exciting vision. Whilst the REALITI system is the first that has been specifically designed to aid tracheal intubation, similar technology has been developed for diagnostic bronchoscopy and lung biopsies. The Ion™ endoluminal system (Intuitive Surgical, Sunnyvale, CA, USA) is a robot-assisted platform for performing minimally invasive lung biopsy procedures. The robotic components allow advanced manoeuvrability and navigation of a catheter into the lung periphery using fibreoptic shape-sensing technology to precisely position and stabilise the catheter for accurate lung biopsies. It also contains software which generates a three-dimensional (3D) reconstruction of the airway using computed tomography (CT) imaging of the patient's lungs, allowing for a pre-planned path to be created. Fielding et al. 5 used this system in their single-arm prospective study and found a 96.6% success rate in reaching and obtaining samples from targeted pulmonary nodules, concluding that the robot-assisted bronchoscope navigated safely to very small peripheral airways. Rojas-Solano et al. 6 used the Robotic Endoscopy System (Auris Surgical Robotics, San Carlos, CA, USA) to assess the technical feasibility and complications of robotic bronchoscopy. The Robotic Endoscopy System consists of four major components, including a robotic endoscope manipulated by two arms under the control of a physician at the operator console. Another key feature is the facilitation of distal navigation of the bronchoscope by the design feature of 180° deflection in any direction. The study showed an absence of serious adverse events and found that the limitations of the conventional transbronchial approach were addressed. This included: the continuous visualisation of suspected lesions; the precise control of the bronchoscope's movements; the locking capability of the bronchoscope; and thorough training of the participants before starting the study. There are other examples of robotic technology being applied to assist bronchoscopy 7, 8, but none yet in the realms of tracheal intubation. These reports, along with the data presented by Biro et al. 3, raise a number of questions relating to feasibility and clinical relevance. For example, if robot-assisted tracheal intubation were to be implemented, would it lead to a significant step change in safety of airway management, or would it merely be an additional tool in the growing list of devices that we may use to incrementally refine management? If this device is highly effective in an automatic mode, will the role of anaesthetists in managing the airway be diminished? How would these devices fare in the difficult airway in which abnormal anatomy or bleeding are present? Would this tool be superior to awake tracheal intubation techniques 9? What would the costs be and would the frequency of failure in tracheal intubation justify the costs? Finally (perhaps the authors may consider the innovation pathway presented by Young 10) could this unpatented product, or a subsequent patented iteration, receive the necessary investment and support from manufactures to eventually bring a product to market? The answer may be gleaned by considering the experience of our surgical colleagues. The application of robotic technology in surgery was first described in the 1980s. The morbidity of open surgical procedures was the driving force for the development of less-invasive approaches. Advances in endoscopy gave rise to laparoscopic surgery and the early experience in robotic surgery reflected a natural evolution of widespread laparoscopic applications. In 1985, the PUMA 560® robot (Advance Research & Robotics, Inc., Oxford, CT, USA) was used to perform the first robotic-assisted CT-guided stereotactic brain biopsy 11. In 1988, the PROBOT® (Imperial College London, London, UK) was used for the first robotic-assisted transurethral resection of the prostate 12. Several surgical robots followed, including the ZEUS® robotic surgical system (Computer Motion, Inc., Santa Barbara, CA, USA). In 2003, Intuitive Surgical, Inc., a California-based surgical robotics company, acquired Computer Motion, Inc. and phased out ZEUS® in favour of the da Vinci® surgical robot 13. This system became the first robotic platform approved by the Food and Drug Administration (FDA) for general laparoscopic surgery. The da Vinci® system constitutes the most commercially successful robotic surgical platform to date. Features such as a dual-channel endoscope allow simultaneous image magnification and 3D immersive visualisation of the target anatomy. Combined with wristed robotic instruments that provide multiplanar resection capability, the result is substantially enhanced surgical dexterity compared with conventional laparoscopic surgery. There have been seven design iterations and over 5000 units sold worldwide. Although originally designed for cardiac surgery, the major initial foothold was in urology for performing prostatectomy. Food and Drug Administration (FDA) approval followed in most surgical specialties and in 2005, head and neck surgery experienced an evolution similar to that of traditional open abdominal and pelvic procedures. Transoral robotic surgery (TORS) was pioneered in the University of Pennsylvania and in 2009, the FDA approved the da Vinci system for resection of cancers of the oropharynx, larynx and hypopharynx 14. In the last decade, TORS has evolved from a proof-of-concept to a standard-of-care in high-volume robotic centres worldwide 15. Although level-1 evidence of effectiveness is controversial, the ever-increasing applications for more precise, less invasive surgery and the associated explosion of the ‘surgical tech’ sector have created a lucrative market worth £4.1 billion that is expected to exceed £18 billion within the next 5 years 16. As patents run out, the monopoly that Intuitive Surgical has thus far enjoyed is being challenged by new players such as the Versius (Cambridge Medical Robotics) and Verb (Google and Johnson and Johnson) systems. Competition is driving down cost and the landscape is shifting; perhaps the term ‘digital surgery’ better describes the next generation of flexible surgical robots. Real-time navigation; instrument miniaturisation; portable, smaller systems; algorithms for data capture, storage, analysis and shared learning; in addition to incorporation of advanced robotic capability better describe an exciting future. Embracing novel technology for patient benefit has been the cornerstone of medical and, in particular, anaesthetic progress. In the last 20 years of peri-operative practice, we have adopted mathematical models to more effectively deliver both total intravenous anaesthesia 17 and automatically titrate volatile anaesthetic; we have incorporated devices analysing cerebral electrical activity to report depth of anaesthesia; we have placed ever-smaller cameras on a variety of devices; and we have integrated ultrasound into countless diagnostic and interventional procedures 18. However, surgeons have harnessed robotic technology in ever-increasing settings, and rather than fear losing their jobs, robots have in fact improved their practice. Put simply, good surgeons have been made better by incorporating robot-assisted technology, and this has potential benefits to patients. A scan around many an operating theatre will reveal that robotic technology is not in fact the future, but indeed the present. The competition in the surgical market for robot-assisted technology is unprecedented. Anaesthesia is often seen as an innovative specialty 10. Important innovations on the horizon include artificial intelligence techniques and machine learning to predict peri-operative hypotension, bispectral index values and postoperative mortality 19. However robotic technology appears to be one area in which we have lagged behind, but in the dynamic landscape of medical technology, the implementation of robot-assisted practical anaesthetic interventions is merely one development that may prove to be an integral factor in improving the quality of peri-operative care. Should the data emerge to support the utility of such novel technologies, it is our duty to be the responsive to this change and embrace it. KE is an Editor of Anaesthesia. No other competing interests declared.

Topics & Concepts