First Human Robotic Tracheal Intubations
We successfully used a robotic intubation system, remotely controlling a standard video laryngoscope using a commercial joystick. Using the KIS, tracheal intubation was successful in 11 out of 12 patients within 100 s; in one patient, fogging prevented a successful KIS use.
Several studies have looked at the learning curve of anaesthesia residents to achieve acceptable success rates when performing tracheal intubation. Kopacz and colleagues stated that a 90.8% of success rate can be achieved after 45 attempts. Konrad and colleagues determined a significantly higher number of attempts of 57 tracheal intubations to achieve success rates of 90%, and de Oliveira Filho determined a pooled success rate of 85% after 43 tracheal intubation attempts. These studies have been performed with anaesthesia residents; however, similar learning curves can be determined for staff experienced in direct laryngoscopy when acquiring skills for using video laryngoscopes for tracheal intubation. Within the limitation of this small-scale pilot study, a success rate of 91% for the first 12 attempts is a clinically very favourable rate. Our study did not show any significant learning curve in terms of success rate because of the already high initial rate, whereas Konrad and colleagues showed a significant learning curve for the success rate during the first 20 attempts reaching ~65% after 20 attempts. Because of the small number of patients in this pilot study, we did not find any significant improvement in the time it took to perform the intubation. A significantly bigger number of patients are needed to evaluate this question.
The Pentax AWS video laryngoscope was specifically chosen for this project since it offers two features: there are two additional ports on the transparent blade, one for suction and the other for attaching a standard tracheal tube; it also features a crosshair, which, once placed in the area of the vocal cords, is supposed to indicate proper positioning of the scope for easy advancement of the tracheal tube into the trachea. In contrast to the other video laryngoscope, such as the widely popular Glidescope (Verathon, USA), there is no need for a stylet, and insertion of the tracheal tube can be achieved by simply pushing the tube forward using the groove on the blade as a rail. Several studies have determined the success rate of intubation with the Pentax video laryngoscope in non-difficult airways and determined success rates between 98% and 100% and mean intubation times of ~20 s. Intubation times were defined as the time to insert the blade into the mouth until passage of the tracheal tube is completed. This definition corresponds to what we defined as 'intubation time' which was 57 s in the present study. In order to put this difference in perspective, one has to take into account the early stage of development of the system, the pilot nature of the study, the limited number of patients, and similar developments of robotic systems in medicine. This time difference is comparable with similar time differences when surgery moved from open techniques to endoscopic and robotic approaches in the early stage of training and development. It is interesting to note that intubation times in the present study were significantly faster than those in a previous mannequin study where fibreoptic intubations were performed using a multipurpose surgical robot.
In one patient, intubation using KIS was not successful because of fogging of the Pentax video laryngoscope. Fogging has been described as rare or more common using this type of video laryngoscope. As recommended by the manufacturer, preheating of the laryngoscope and application of antifogging liquid was performed before each intubation attempt in this study. Another limitation of this type of laryngoscope is that the maximum width of the blade is 2.5 cm; this might create problems in patients with limited mouth opening, who in this pilot study were excluded.
The controller used by the KIS is a standard gaming joystick with the possibility to programme up to 12 buttons and five axles. All movements carried out by the wrist or arm of a human being when holding the laryngoscope can be simulated. The operator (T.M.H.) had familiarized himself with these movements and operational modes of the joystick in extensive mannequin training. At present, there is no force feedback integrated into the system and all movements need to be directed by a human being using the joystick. However, the ultimate goal is to fully automate the intubation process. The first step towards this goal is to automate the navigation of the blade of the video laryngoscope through the mouth and pharynx, avoiding damage to the teeth and tissue. At this point, the operator could easily align the video laryngoscope with the trachea and manually insert the tube. The KIS is designed to also allow for remote intubations and could be used as part of a telemedical system.
There are several limitations to this study. This study was designed as a pilot study focusing on measuring the success rate and intubation times in a small number of patients. More patients are needed to evaluate its performance and safety. The KIS is a prototype with future technical focus being placed on making the system smaller, portable, and semiautomatic.
We present the first robotic intubation system for oral tracheal intubation tested in humans. These first tests revealed a good success rate within reasonable time frames. Future studies are needed to evaluate the performance and safety of the system.
Discussion
We successfully used a robotic intubation system, remotely controlling a standard video laryngoscope using a commercial joystick. Using the KIS, tracheal intubation was successful in 11 out of 12 patients within 100 s; in one patient, fogging prevented a successful KIS use.
Several studies have looked at the learning curve of anaesthesia residents to achieve acceptable success rates when performing tracheal intubation. Kopacz and colleagues stated that a 90.8% of success rate can be achieved after 45 attempts. Konrad and colleagues determined a significantly higher number of attempts of 57 tracheal intubations to achieve success rates of 90%, and de Oliveira Filho determined a pooled success rate of 85% after 43 tracheal intubation attempts. These studies have been performed with anaesthesia residents; however, similar learning curves can be determined for staff experienced in direct laryngoscopy when acquiring skills for using video laryngoscopes for tracheal intubation. Within the limitation of this small-scale pilot study, a success rate of 91% for the first 12 attempts is a clinically very favourable rate. Our study did not show any significant learning curve in terms of success rate because of the already high initial rate, whereas Konrad and colleagues showed a significant learning curve for the success rate during the first 20 attempts reaching ~65% after 20 attempts. Because of the small number of patients in this pilot study, we did not find any significant improvement in the time it took to perform the intubation. A significantly bigger number of patients are needed to evaluate this question.
The Pentax AWS video laryngoscope was specifically chosen for this project since it offers two features: there are two additional ports on the transparent blade, one for suction and the other for attaching a standard tracheal tube; it also features a crosshair, which, once placed in the area of the vocal cords, is supposed to indicate proper positioning of the scope for easy advancement of the tracheal tube into the trachea. In contrast to the other video laryngoscope, such as the widely popular Glidescope (Verathon, USA), there is no need for a stylet, and insertion of the tracheal tube can be achieved by simply pushing the tube forward using the groove on the blade as a rail. Several studies have determined the success rate of intubation with the Pentax video laryngoscope in non-difficult airways and determined success rates between 98% and 100% and mean intubation times of ~20 s. Intubation times were defined as the time to insert the blade into the mouth until passage of the tracheal tube is completed. This definition corresponds to what we defined as 'intubation time' which was 57 s in the present study. In order to put this difference in perspective, one has to take into account the early stage of development of the system, the pilot nature of the study, the limited number of patients, and similar developments of robotic systems in medicine. This time difference is comparable with similar time differences when surgery moved from open techniques to endoscopic and robotic approaches in the early stage of training and development. It is interesting to note that intubation times in the present study were significantly faster than those in a previous mannequin study where fibreoptic intubations were performed using a multipurpose surgical robot.
In one patient, intubation using KIS was not successful because of fogging of the Pentax video laryngoscope. Fogging has been described as rare or more common using this type of video laryngoscope. As recommended by the manufacturer, preheating of the laryngoscope and application of antifogging liquid was performed before each intubation attempt in this study. Another limitation of this type of laryngoscope is that the maximum width of the blade is 2.5 cm; this might create problems in patients with limited mouth opening, who in this pilot study were excluded.
The controller used by the KIS is a standard gaming joystick with the possibility to programme up to 12 buttons and five axles. All movements carried out by the wrist or arm of a human being when holding the laryngoscope can be simulated. The operator (T.M.H.) had familiarized himself with these movements and operational modes of the joystick in extensive mannequin training. At present, there is no force feedback integrated into the system and all movements need to be directed by a human being using the joystick. However, the ultimate goal is to fully automate the intubation process. The first step towards this goal is to automate the navigation of the blade of the video laryngoscope through the mouth and pharynx, avoiding damage to the teeth and tissue. At this point, the operator could easily align the video laryngoscope with the trachea and manually insert the tube. The KIS is designed to also allow for remote intubations and could be used as part of a telemedical system.
There are several limitations to this study. This study was designed as a pilot study focusing on measuring the success rate and intubation times in a small number of patients. More patients are needed to evaluate its performance and safety. The KIS is a prototype with future technical focus being placed on making the system smaller, portable, and semiautomatic.
We present the first robotic intubation system for oral tracheal intubation tested in humans. These first tests revealed a good success rate within reasonable time frames. Future studies are needed to evaluate the performance and safety of the system.
Source...