Angulated instruments demonstrating their potential in encircling structures.
First use of the angulated instrument in experimental work.
The surgical manipulator ARTEMIS (advanced robotic and telemanipulation for minimally invasive surgery; Forschungszentrum Karlsruhe, Karlsruhe, Germany): the 3-dimensional simulation of surgical tasks and movements of the manipulator system.
The surgical manipulator ARTEMIS (advanced robotic and telemanipulation for minimally invasive surgery; Forschungszentrum Karlsruhe, Karlsruhe, Germany): the full-scale functional model ready for experimental application.
The positioning system TISKA (trocar and instrument positioning system Karlsruhe; Karl Storz GmbH & Co, Tuttlingen, Germany): a demonstration of the invariant point.
The positioning system TISKA (trocar and instrument positioning system Karlsruhe; Karl Storz GmbH & Co, Tuttlingen, Germany) and the motor-driven camera-guiding system FIPS (flexible instrument positioning system; Karl Storz GmbH & Co).
Evaluation of solo surgery systems. Results of conducting surgery with ENDOASSIST (Armstrong Healthcare Ltd, High Wycombe, England); AESOP 2000 (Computer Motion, Goleta, Calif); and FIPS (flexible instrument positioning system; Karl Storz GmbH & Co, Tuttlingen, Germany) voice-controlled, motor-driven, camera-guiding systems.
Forceps with the integrated sensors.
The shadow optic.
Buess GF, Schurr MO, Fischer SC. Robotics and Allied Technologies in Endoscopic Surgery. Arch Surg. 2000;135(2):229–235. doi:10.1001/archsurg.135.2.229
Endoscopic surgery was developed in the 1970s and 1980s, with initial work conducted by pioneering surgeons. After the development of laparoscopic cholecystectomy, the breakthrough of endoscopic surgery had a great effect on all surgical specialties. Starting with rather simple procedures, such as cholecystectomy, a rapid progression toward more complex procedures, such as reflux or colonic surgery, took place. It was realized at this time that the existing endoscopic instruments allowed only a limited preciseness when performing the procedures, and part of the information from inside the abdominal cavity was not available to the surgeon. This prompted a discussion with engineers concerning the development of more advanced technologies to give those performing endoscopic surgery the same quality of information and manipulation that surgeons have when performing open surgery. These qualities include (1) instruments and manipulators that allow surgical action under endoscopic control with all degrees of freedom; (2) devices that provide surgeons with tactile feedback; and (3) vision systems that provide surgeons with the same quality of visual information as with open surgery, namely, high resolution, excellent color quality, precise spatial information, and a constant clear view for optimal surgical action. At the end of 1999, some of the aforementioned quality concepts found their way into the surgical routine, but most of the concepts are still being developed. Another decade will pass before endoscopic surgery procedures will be closer to the technological goals.
Until the 1980s, technological devices used for performing surgical tasks have been simple, standard instrumentations. The introduction of endoscopic surgery had a great effect on surgery conditions in different fields. The surgeon could now use far better tools for one of the most important medical tasks: the reduction of invasiveness (eg, in the course of the removal of organs). At the same time, surgeons experienced how complex the implementation of new technologies into the surgical routine can be.
Decades of development work by pioneers in surgery had been done before laparoscopic cholecystectomies were performed. In 1954, Raimund Wittmoser, MD, PhD,1 a surgeon from Germany, developed the first thoracoscopic procedures for sympathectomy. In 1971, Kurt Semm, MD, PhD,2 developed important laparoscopic techniques, and one of the first complex operations on the gastrointestinal tract, the full-thickness resection of rectal tumors, was introduced by one of us (G.F.B.).3
Based on this pioneering work, which has an extensive tradition in Germany, we installed the programs for the future development of endoscopic surgery in 1989 at the university clinic in Tübingen, Germany. Development programs started in cooperation with one of the most advanced national research centers in Germany, the Forschungszentrum Karlsruhe (FZK) in Karlsruhe. At the same time, educational programs for endoscopic surgery training started at our institution.
Advanced procedures have been rapidly developed and introduced into surgical practice, prompted by a close cooperation between our university and Alfred Cuschieri, MD, PhD, from Dundee, Scotland.2 At the same time, discussions began regarding the manner in which technology should move forward to ensure better and safer advanced surgery.
Three development aims have been defined by FZK and us, which fulfilled the goals of the development work until 1996. The basic aims for developing advanced technologies as defined in 1989 are as follows: (1) Better manipulation of endoscopic instruments, enhancing the degrees of freedom of movement. Owing to the fixation of surgical instruments at the port sides, the movements of the tips of the instruments are severely restricted. To compensate for these restrictions, surgeons could use curved4 and bayonet-formed3 instruments. The development of surgical instruments with active bending possibilities and active rotation of the heads was the next important step, followed by the development of manipulator systems for ergonomic control with the use of all degrees of freedom (advanced robotic and telemanipulation for minimally invasive surgery [ARTEMIS; FZK, Karlsruhe, Germany]). (2) The development of mechanisms to provide the surgeon with tactile information (the first prototypes of tactile sensors). (3) The development of vision systems with the aim of providing a constant clear view, better spatial information, and better resolution, as well as a complete overview during the operation.
In open surgery, a needle holder manipulated by the human angulation system—wrist, elbow, and shoulder—allows the surgeon to guide the tip of the needle through each desired angle and rotation. The implementation of this device in an endoscopic instrument needs to include an angulation system inside the respective cavity. This can be achieved by a combination of joints and a mechanism for rotating the tip. The first instrument designed to achieve this was developed jointly by FZK and us. Angulation was achieved by a series of segments, designed by the engineer Ing Jürgen Müglitz, PhD.5Figure 1 shows one of these instruments and the potential it has to encircle structures. In clinical use the instrument can encircle the esophagus or the colon, for instance.
Because of the complex steering of this instrument, electrical drives have been integrated. The instrument was used for the first time in the early 1990s in experimental work (Figure 2). During the application of this active angulating and rotating instrument, we could see that the handling was difficult and nonintuitive. The decision was made to implement a surgical manipulator so that in the future, surgical actions could be performed intuitively. Manipulators have been used before (eg, for the manipulation of nuclear substances in spaces human beings could not enter). These basic developments have been transferred into the first experimental surgical manipulator.
The goals of implementing surgical manipulators into surgery are (1) to enhance manipulation capabilities and (2) to enhance performance precision. It could already be demonstrated in other fields of surgery that by implementing robots and robotic manipulators, the preciseness of surgical actions could be dramatically enhanced. For example, robots used for performing hip replacements include Robodoc (Integrated Surgical Systems Inc, Sacramento, Calif) and CASPAR (computer-assisted surgical planning and robotics; orto MAQUET, Rastatt, Germany).6 Another goal is (3) to replace human power. This has not been the focus of interest from the beginning, but spin-off solutions of the robotic manipulator, such as a motor-driven camera-guiding system (flexible instrument positioning system [FIPS]; Karl Storz GmbH & Co, Tuttlingen, Germany) or a positioning system (trocar and instrument positioning system Karlsruhe [TISKA]; Karl Storz GmbH & Co)7 and AESOP 20008 (Computer Motion, Goleta, Calif), which was the first camera-guiding system on the market, can be used today to reduce the number of surgeons and assistants necessary at the operating table.
As a first step in the development work for the surgical manipulator ARTEMIS, a simulation program was designed that allowed for the 3-dimensional simulation of all surgical tasks and manipulator system movements. Figure 3 shows ARTEMIS, designed in 1992, and the 3-dimensional graphics. In 1995, a full-scale functional model was ready for experimental application (Figure 4). The instrument shown in Figure 1, which maintains all degrees of freedom, was implemented into this surgical manipulator system. Steering was performed by the tip-tracking technique.
When using modern technologies, such as surgical manipulators or camera-guiding systems, the question of how to command them is of great importance. The surgeon should be able to operate the machine in the most intuitive way possible. The information should be given to the machine in a highly reliable and safe manner, without bringing stress to the surgeon. The input device has to follow intuitive, self-explaining principles.
Therefore, we developed an input device (slave manipulator) following the principle used today also by intuitive surgery. The surgeon sits in an ergonomic position with an optimal view of the monitor and guides the input devices with both hands, manually steering the tip of the instrument by the tip-tracking technique. The surgeon is actually guiding a mechanism similar to the angulated instrument by intuitively performing the movements at the tip of the mechanism. The electronic system transforms the surgeon's movements into electronic commands that directly steer the operating instrument. This system was applied for the first time in 1996 in experimental animal work. It could be clearly shown that intuitive surgical work in complex operations, such as surgery on the distal esophagus, became possible. Because of the complexity of the technology, the project was not continued by FZK and us, and no industrial partner was found for the realization of this high-risk venture.
There is a clear separation between the tasks of the surgeon and those of the assistant in endoscopic surgery. The assistant's task is restricted to guiding the camera or retracting organs. The surgeon's tasks, such as dissection and ligation of vessels or suturing, are only performed by the surgeon. This clear separation of tasks during the operation, and the surgeon's dependence on the camera position as given by the camera assistant, was a strong argument for the integration of camera-guiding or retracting systems that could be manipulated by the surgeon.
The first major development in this area was conducted by Computer Motion. From the beginning, AESOP 20008 used the principles of robotic manipulators9 known from other technological areas. Intensive work has been done focusing on the man-machine interface. Besides the construction of a robotic manipulator for optimal camera guidance, a satisfactory system had to be developed to allow the surgeon to intuitively control the camera to reach the desired position as quickly as possible. Computer Motion first focused on hand and foot pedals for steering the camera, but the hand pedal was not practical because it took the surgeon's hand away from the operation, and the foot pedals were not ergonomic. Speech control was therefore the focus of interest. By specific voice commands, the camera could be moved to the desired position.
After the first experiment using the surgical manipulator ARTEMIS,10 we focused on developing camera-guiding and retracting systems together with FZK. From the beginning, our interpretation of solo surgery was that both aspects, camera guidance and surgical retraction, have to be supported by adequate new technologies.
The first device developed by FZK and us was the TISKA positioning system.7 This system uses parallelograms so that whatever position the surgical instrument or camera is in, the invariant point, where the trocar cannula goes through the abdominal wall, is stable (Figure 5). A magnetic brake is linked to the TISKA system so that by pressing a button, the brake is released, and the instrument or endoscope can be guided to the new position by one hand. Positioning systems that do not have an invariant point usually need to be repositioned with 2 hands. Manual control is preferred because positions do not change frequently during surgery; the surgeon is not considerably distracted from surgical tasks, and manually controlled instruments are easier to maneuver and safer than motor-driven instruments.
The principle of this camera-guiding system was already developed for experimental work (the surgical manipulator ARTEMIS). By transferring technology to other applications (spin-off solutions), we continued to develop a system to apply to everyday endoscopic surgery. The FIPS11 system maintains the invariant point at the abdominal wall, and mechanical devices allow movement around the invariant point without the need for complex electronic calculation, as, for example, with the AESOP 20008 system. Motors are integrated for all movements. For the first time, the rotation of the angulated endoscope was integrated into the camera-guiding system. We believe that camera-guiding systems should also be applicable to more complex surgical tasks so that the use of an angulated endoscope and rotation is imperative. Figure 6 shows the solo surgery setup, uniting the FIPS and TISKA systems.
The positioning and camera-guiding systems have been further developed in work conducted jointly by Karl Storz GmbH & Co, one of the major companies in the field of technology for endoscopic surgery, FZK, and us. Evaluation of the systems for solo surgery was performed in an experimental setting. Evaluation in a clinical setting often does not allow systematic work because comparable situations are rarely reproduced. For this evaluation, we integrated a pig's liver with the gallbladder into the anatomical plastic model we use in our training center and performed the operations in our experimental operating theater. The experiments were mainly performed by Alberto Arezzo, MD, PhD, and other surgeons from our group. The results of this trial are shown in Figure 7.
With the main information from this trial, it was concluded that the systems that are applicable today can fulfill the demand for rapid and ergonomic surgical work. The time needed to prepare the system before the operation and take it down afterwards is still notable. The systems today are not considerably timesaving in the operating theater, but they do allow a reduction in the number of assistants.
Parallel to the surgical manipulator ARTEMIS and solo surgery systems, we focused on developing an operating room system. This methodically controls the different subsystems (eg, the insufflation unit or the camera). In today's operations, these subsystems are independent, stand-alone units, making central control of the different functions and ergonomic application impossible.
Together with Dornier Medizintechnik GmbH, Munich, Germany, we developed the first operating room system for endoscopic surgery. This system has the following features: (1) The integration of all subsystems equipped with a computer interface necessary for performing endoscopic surgery. (2) The integration of a central steering computer for the control of all functions. (3) An arm covered by a sterile drape reaching into the sterile field. (4) A flat-screen monitor with a sterile cover inside the sterile field with self-explanatory input possibilities for adapting the subsystems to the surgeon's needs. (5) The possibility to program the system in such a way that the individual adjustment of all subsystems for the endoscopic operation of a gallbladder, for example, can be programmed by one touch.
Following the development of the operating room system for endoscopic surgery, major companies on an international basis focusing on advanced technology started development work on similar operating room systems. Today, Computer Motion offers a system in cooperation with other companies. Siemens AG in Erlangen, Germany, worked jointly with several German and international companies to link different subsystems together in an open electronic architecture for optimal control. Storz as well as Olympus Winter & Ibe GmbH, Hamburg, Germany, also offer subsystems that use a computer-controlled system concept.
For the future, it will be important that systems allow an open architecture so that new developments (eg, ultrasonically activated dissectors) can be easily integrated. Only these systems will guarantee that all necessary equipment in the operating room can be linked together for central control and steering.12
Two different aspects of tactile information have to be considered. The first is that tactile information gives the surgeon an idea of how firmly the tissue is grasped. This also gives indirect information about the elasticity and the firmness of the tissue. In technical terms this aspect of tactile information is called force feedback. This means that the force applied to the tissue is fed back through the input device to the surgeon's hands.
Another important piece of information that is lost when performing endoscopic surgery is the information about elasticity and firmness of structures, which, for example, is necessary for locating a tumor in the colon and distinguishing the margins of the tumor from the surrounding organs. In this context, we performed some basic development work together with research and development groups from Dornier. A specific forceps was developed for the palpation of organs. Microstructure sensors have been integrated inside these forceps (Figure 8). In experimental work, we have been successful in graphically demonstrating the tactile information on the monitor; however, the microstructure sensors do not provide reliable and reproducible information, so use in the clinical routine is still not possible.13
Parallel to the development of microstructure sensors, we focused on developing a vibrotactile sensor. The function of the vibrotactile sensor14 is based on the following physical principle: tissue stiffness is determined by its reaction to vibration. The tip of the vibrotactile sensor is connected to the tissue to be examined. Vibration is applied to the tissue, and different tissue properties are determined by the resonance range. Normal soft tissue has a resonance of 16 Hz; bony structures, 240 Hz. A prototype of the vibrotactile sensor has been applied clinically in head and neck surgery.15
In all areas of endoscopic surgery, the quality of vision systems strongly influences the preciseness and safety of the procedure. In context with the application of robotics, which allows more precise surgical manipulation, the quality of vision systems is even more decisive. The development of the robotic manipulator (Intuitive Surgical Inc, Mountain View, Calif) highlights this point with the integration of a 3-dimensional vision system.
In our group, we have a long tradition concerning aspects of enhancing the quality of vision. The first important development is the stereoscopic endoscope for endorectal surgery. This endoscope16 has 2 integrated Rodlens optics, one for the left eye and one for the right. The optical information is, according to our experience, the best spatial information available. The high resolution of the Rodlens optic is guided directly to the eye so that there is no loss of image quality.
In the beginning of the 1990s, we focused, along with FZK, on the development of a stereoscopic endoscope using shutter glasses and monitor technologies. The evaluation of these devices showed that there was clearly better spatial information, which was documented in experimental trials.17 Nevertheless, the application of stereoscopic video endoscopes could never achieve clinical relevance because stereoscopic video techniques lose light and resolution so that factors important for a precise endoscopic view are negatively influenced. Experimental and clinical trials using high-resolution camera techniques could demonstrate that high resolution and brilliant color quality guarantee important spatial information.
A simple solution to achieve better spatial information was recently developed. We integrated 2 different light sources into a standard endoscope (shadow optic)18 as shown in Figure 9. This light distribution creates more precise spatial information by integrating shadows into the field of endoscopic vision. Other aspects that optimally serve the needs of robotic surgery should be integrated into future endoscopes: heating of the optical instrument guarantees a fogless lens during the whole procedure; rinsing makes it possible to clean the endoscope when the view is impaired; and insufflation of fresh gas to the tip of the endoscope prevents particles, created by high frequency or ultrasonically activated dissectors, from forming a layer on the surface of the front lens.
Surgery has been a low-technology discipline for a long time. Simply designed instruments such as needle holders and graspers have been used for centuries to perform different surgical tasks. During the introduction in a session (T. Scherstein, oral communication, November 14-15, 1966) focusing on modern, high-technology surgery, the presenter showed 4 slides to compare the adaptation of modern technologies in different fields. The first slide showed a train in 1895; the second, a train in 1995. The third slide showed surgical instruments in 1895; the fourth, surgical instruments in 1995. No notable differences could be seen between the third and fourth slides.
This situation clearly demonstrates that surgeons have not kept up with the developments occurring in other sectors of science, where modern technology has deeply influenced scientific progress. Only with the introduction of endoscopic surgery has rapid adaptation of modern technologies taken place, such as the integration of video technologies or electronic instrument control into the surgeons' routine work. The adaptation of robotics and robotic manipulators is the culmination of the adaptation of technology.
Most of the principles of robotics have not specifically been designed to be applied to surgery, but developments over the decades in the area of nuclear substance manipulation, for example, have been transferred from their primary use to be applied to surgery. Technology transfer is dominating the development process in many fields. These principles of technology transfer explain why the development process can happen in such a rapid way. Robotic manipulation is widely enhancing the potential of endoscopic surgery. It has already been demonstrated in cardiac surgery that operations that could not have been performed endoscopically by basic technology using rigid instruments can now be mastered.
The adaptation of highly complex robotic systems also represents the considerable difficulties concerning the working process inside clinics. New operating theaters have to be designed to allow the integration of space-consuming robotic systems as part of a complete operating room system, equipping the theater with all the technical needs for endoscopic surgery. The application of robotic systems needs extensive financial investment, which can only be provided by highly specialized clinics at which a lot of expensive specialty operations are performed. At the same time, the staff handling this complex technology is not available in standard operating departments. Surgeons have to be specifically educated in the use of modern technologies, nurses have to be trained to perform the assisting tasks, and the most complex devices, such as the intuitive surgical manipulators, have to be controlled by engineers available inside operating theaters.
Future development will show whether the afore-mentioned complex conditions will allow these robotic manipulators to be integrated on a wide scale of endoscopic procedures. We believe that more simply designed and easier to apply surgical devices that maintain the necessary degrees of freedom will be developed. During the next years, the task of engineers is to make these highly complex technologies simpler, cheaper, and more intuitive for application by surgeons under more standardized conditions.
A simpler spin-off solution for today's routine use is currently being developed. Camera-guiding and instrument-positioning systems are designed to allow more intuitive surgical work and to compensate for the surgeon performing operations without having the field of vision under direct command. Camera-guiding systems with intuitive, self-explanatory steering concepts will allow surgeons to be in better control of their tasks. In this context research work has to focus on developing easily understandable input devices to control the complex technology in self-explanatory ways. The technology for camera-guiding and instrument-positioning systems enables the surgeon to leave the standard team environment behind, leading to operations in which a single surgeon is performing all of the necessary tasks.
The technology available today will only be effectively integrated into routine surgery inside highly specialized centers, at which only a small number of variations of procedures are performed. For application to a wider scale of procedures, systems for controlling operating-room technology will need to be much more flexible. These operating room systems will be a necessity in a few years, when high-technology operating theaters will be equipped for operations that use complex technologies. The stand-alone units used today often do not allow optimal adjustment and control of important functions. Therefore, the chaotic organization of today's operating theaters will not provide the best safety when modern technology is integrated.
Currently, tactile information is lost in endoscopic surgery. This loss is not of great importance with simple procedures such as cholecystectomies, but more complex procedures such as tumor surgery have extensive restrictions without the possibility of defining the tumor location and margins. In this situation, the surgeon is confronted with restrictions never faced in open surgery.
In endoscopic surgery the quality of each step of the operation depends on the quality of vision. The vision systems have to match the quality that is offered to the human brain by the complex and still the best vision system available: the human eye. An important task for those conducting research and development in the future will be to provide the surgeons using endoscopes with the same high-resolution vision that the brain receives when no endoscopes and video systems are used. The potential of keeping the endoscopes fog-free and easy to clean has to be transferred from the physiologic characteristics of the human eye to the technological solution, and spatial resolution has to be reintegrated into future images with high-definition, high-contrast, high-color trueness.
Corresponding author: Gerhard F. Buess, MD, PhD, Zentrum für medizinische Forschung, Sektion für MIC, Waldhörnlestr 22, 72072 Tübingen, Germany (e-mail: firstname.lastname@example.org).