Telerobotic Laparoscopic Cholecystectomy: Initial Clinical Experience With 25 Patients

Abstract

Objective

To determine the safety and feasibility of performing telerobotic laparoscopic cholecystectomies. This will serve as a preliminary step toward the integration of computer-rendered three-dimensional preoperative imaging studies of anatomy and pathology onto the patient’s own anatomy during surgery.

Summary Background Data

Computer-assisted surgery (CAS) increases the surgeon’s dexterity and precision during minimally invasive surgery, especially when using microinstruments. Clinical trials have shown the improved microsurgical precision afforded by CAS in the minimally invasive setting in cardiac and gynecologic surgery. Future applications would allow integration of preoperative data and augmented-reality simulation onto the actual procedure.

Methods

Beginning in September 1999, CAS was used to perform cholecystectomies on 25 patients at a single medical center in this nonrandomized, prospective study. The operations were performed by one of two surgeons who had previous laboratory experience using the computer interface. The entire dissection was performed by the surgeon, who remained at a distance from the patient but in the same operating room. The operation was evaluated according to time of dissection, time of assembly/disassembly of robot, complications, immediate postoperative course, and short-term follow-up.

Results

Twenty of the 25 patients had symptomatic cholelithiasis, 1 had a gallbladder polyp, and 4 had acute cholecystitis. Twenty-four of the 25 laparoscopic cholecystectomies were successfully completed by CAS. There was one conversion to conventional laparoscopic cholecystectomy. Set-up and takedown of the robotic arms took a median of 18 minutes. The median operative time for dissection and the overall operative time were 25 and 108 minutes, respectively. There were no intraoperative complications. There was one postoperative complication of a suspected pulmonary embolus, which was treated with anticoagulation. All patients were tolerating diet at discharge.

Conclusions

Laparoscopic cholecystectomy performed by CAS is safe and feasible, with operative times and patient recovery similar to those of conventional laparoscopy. At present, CAS cholecystectomy offers no obvious advantages to patients, but the potential advantages of CAS lie in its ability to convert the surgical act into digitized data. This digitized format can then interface with other forms of digitized data, such as pre- or intraoperative imaging studies, or be transmitted over a distance. This has the potential to revolutionize the way surgery is performed.

Computers and technology are increasingly interacting with surgeons both inside and outside of the operating room, as exemplified by the rapid adoption of laparoscopy into routine use. The computer’s ability to enhance, modify, or transform electronic data is changing patient management before, during, and after surgery. As such, these technologic advancements are having an ever-increasing influence on the way surgery is planned and performed. ^1–5

We and others have previously reported preoperative liver imaging studies that show computer translation of conventional computed tomography scans into three-dimensional virtual reality images that sharply delineate the patient’s anatomical structures as well as the presence of any pathology. ^6–8 By using this model, the surgeon can choose dissection planes before and during surgery. This significant advance in imaging technology has led to image-guided surgery in specialties such as neurosurgery and has led to an increased interest in computer-assisted surgery (CAS). ^9,10 To gain intraoperative benefits from virtual reality imaging, the entire surgical operation must be transformed into a form of communication that can be integrated with the information transmitted by real-time imaging. This is accomplished by transforming the movements of the surgeon’s instruments into electronic signals, or digitization, a form of communication that can be processed by a surgical robotics computer system. As an initial step in this evolution in surgical planning and intraoperative interaction of surgeon with the computer, the present study was designed to determine the feasibility, safety, and utility of performing laparoscopic cholecystectomies with the computer–surgeon interface of a surgical robotic system. The robotic system duplicates the motion of the laparoscopic instruments controlled by the surgeon, who is positioned at a distance from the patient; using this system, laparoscopic cholecystectomies were performed in both acute and nonacute clinical situations. Establishing the safety of performing robotic procedures on abdominal internal organs will pave the way for future definitive studies of computer integration of preoperative imaging studies with real-time, computer-assisted surgery.

CAS offers several advantages: the increase in three-dimensional accuracy, the reproducibility of repeated procedures, the increased precision of movements, and the unique ability to perform surgery over a distance. ¹¹ The robotics systems currently in use, including the system described here, operate on a master/slave principle that involves the use of a “master” computer console operated by the surgeon that integrates computer input, manipulation of the master instruments that control the endoscopic instruments, and the “slave” instruments that are attached to and manipulated by the robotic control arms. ^12–15 This is considered computer-assisted “passive” robotic manipulation because the robot is under the direct control of an operating surgeon. Therefore, the robot does not replace but rather augments the skills of the surgeon.

As an initial clinical study on robotic-aided surgery to establish safety and feasibility, further aims of this study were to determine the practicality of the use of the robot by operating room support staff and surgical staff; location of ports for optimal placement of robotic-controlled instruments to optimize visualization and to minimize opposition of the instruments; and finally to compare postoperative course and patient outcomes with literature standards for laparoscopic cholecystectomies.

METHODS

The use of the surgical robotic system Zeus (ComputerMotion, Galeta, CA) was approved by the European Commission and conformed to all its specifications, including patient consent. The cholecystectomies were performed by one of two surgeons at the same medical center between September 1999 and June 2000. Both surgeons are experienced laparoscopists and had two to five training sessions using the robot on a porcine model in the laboratory before advancing to the clinical study. Patients were assigned in a nonrandomized fashion to have the cholecystectomy performed with the surgical robot. Exclusion criteria included previous upper abdominal surgery, severe comorbid disease and/or American Society of Anesthesiologists (ASA) score of 3, persistent fever or leukocytosis after treatment with antibiotics, abnormal coagulation profile, or concurrent presence of pancreatitis. In each operation, a standard, scrubbed operating team consisting of one surgeon and/or one assistant and a nurse assisted with port placement. The surgical team remained present throughout the entire surgical procedure. Three interactive robotic arms were attached to the siderails of the operating table; two robotic arms manipulated the instruments under the direct control of the surgeon and one arm (“Aesop”) positioned the endoscope by voice command. The operating surgeon was approximately 4 m distant from the patient in the same operating suite and operated by means of the computer console, with telemanipulation of the instruments (Figs. 1–3).

Figure 1. Operating room set-up of robotic surgical system.

Figure 2. Computer–surgeon interface with Zeus.

Figure 3. Telerobotic system (Zeus) in the operating room.

After induction of anesthesia, a pneumoperitoneum was created by a Veress needle. One 10-mm port was positioned periumbilically by the surgeon and used for placement of the endoscope. The endoscopic port was placed 7 to 10 cm distant from the costal margin starting from the estimated position of the gallbladder and in line with the umbilicus. Under direct vision, three additional 5-mm ports were placed, one in the right anterior axillary line in the upper quadrant and two in the left upper quadrant. We determined the optimal placement of the operating ports according to the ability to keep the instruments at an angle between 60° and 90° to each other. The placement of these ports was based on an arcuate line created centered on the estimated position of the gallbladder (Fig. 4). The left operating port was placed at a right angle to the endoscope along this line. The right-sided port was positioned along the arcuate line, 7 to 10 cm distant from the endoscope and 10 to 15 cm distant from the left port to decrease interference among the robotic arms. The telemanipulated 3-mm instruments were inserted through the right upper quadrant and medial left upper quadrant ports. The lateral left upper quadrant port was accessed by the surgical assistant and was used for retraction and clip placement. Dissection of the cystic artery, cystic duct, and gallbladder was achieved by telemanipulation of the hook and grasper instruments on the robotic arms. The assistant placed clips on the cystic artery and duct. Further dissection continued by telemanipulated instruments. Once fully dissected, the gallbladder was removed through the umbilical port in an endobag.

Figure 4. Placement of trocars.

RESULTS

Cholecystectomies were successfully performed on 24 of 25 patients using the Zeus system. Patient characteristics are given in Table 1. Twenty patients had a history of symptomatic cholelithiasis, and the presence of gallstones was confirmed by ultrasound; one patient had a gallbladder polyp; and four patients had acute cholecystitis.

Table 1. PATIENT CHARACTERISTICS

The median time for dissection was 25 minutes (range 14–109). The procedure that took 109 minutes was in a patient with acute cholecystitis. The median total time for set-up and takedown of the robotic arms was 18 minutes (range 13–27). There was one conversion to a conventional laparoscopic procedure in a patient with acute cholecystitis. This procedure was successfully completed by the laparoscopic approach. By evaluating the number of times the robotic arms indicated an interference between arms, we determined that the best placement of ports was with the right port 8 cm distant from the endoscope, and the left lower port in a line at a right angle to the endoscope and at least 14 cm distant from the right port.

An engineer was present for 22 of the procedures, and in 3 cases minor technical adjustments were made to the robotic equipment (in 2 cases resulting from a nonfunctioning grasper and in 1 case resulting from a malfunctioning robotic arm sensor). In each of these instances, once adjustments were made, the operation was successfully completed with the use of the robot.

Twenty-four patients tolerated liquids in the immediate postoperative period; one patient had prolonged nausea for 24 hours. Mean postoperative hospital stay was 3 days. Postoperative complications included one case of suspected pulmonary embolus (angio-enhanced computed tomography results were indeterminate). This patient had a previous history of deep venous thrombosis and therefore had a high risk for such a complication. The patient was given anticoagulation, which was continued after surgery. There were no appreciable complications directly resulting from the use of the robot.

Follow-up at 1 week and 1 month showed one patient with symptoms of reflux disease that had been present before surgery; this patient responded to medical treatment. One patient reported upper abdominal wall pain at a site distant from the port insertion site; it responded to conservative treatment. All patients could tolerate a regular diet and resume normal activity levels.

DISCUSSION

Initial clinical trials using robotics in the operating room have shown the ability of the system to enhance the skill of the surgeon to perform technically precise suturing and dissection. ^14–23 By enhancing the precision of the surgeon, the robot has aided in the development of microsurgical procedures, such as those used in cardiac and infertility surgery, and their advance into the field of endoscopic surgery. The computer interface helps the surgeon perform the microanastomoses using a minimally invasive approach, with all the advantages to the patient of such techniques, including reduced recovery time and better cosmesis.

Our institute is developing the intraoperative use of the surgeon–computer interface with the goal of integrating preoperative imaging studies directly onto organs during surgery to help surgeons plan and perform the dissection and locate pathology. Our initial three-dimensional imaging work was on the liver, using a computer-rendered version of the patient’s imaging studies to locate vascular and ductal anatomy as well as tumor pathology. To accomplish this integration of the surgical act with imaging studies, we are performing two concurrent phases of a study analyzing the use of computer-assisted surgery: the first is in the experimental laboratory performing CAS hepatectomies, and the second is in the operating room with preliminary clinical studies. The clinical phase of this study is reported here. Our aim was to confirm the safety and feasibility of the routine use of robotics in the performance of a laparoscopic technique. Cholecystectomy was chosen as a model because the laparoscopic approach for this procedure is considered the gold standard, and its use is relatively standardized throughout the world; this allows straightforward and direct comparison between computer-assisted and conventional laparoscopic technique.

Our initial experience using the computer-robot interface in performing laparoscopic cholecystectomies in both acute and nonacute settings yielded an operative time that compared favorably with that published in the literature. ^24,25 The one conversion to conventional laparoscopy was in a patient with acute cholecystitis. In this case, the 3-mm instruments were ineffective in grasping the edematous gallbladder, and therefore the anatomy could not be clearly delineated. The Zeus surgical robotic system was originally designed for use in cardiac surgery, and in our trial we had to rely on these finer instruments designed for cardiac microsurgery. As the use of robotics in digestive surgery increases, modifications to the instruments will be necessary; indeed, these are already in the development stage. The average postoperative stay of 3 days is standard at our medical center in France and should not be misinterpreted as resulting from postoperative complications. Overall, these data show that laparoscopic cholecystectomy can be safely and effectively performed using the telemanipulative surgical interface.

The Zeus system is only one of several computer-assisted robotic systems in clinical trials. Another system, Da Vinci (Surgical Intuitive, Mountain View, CA), has also been used clinically to perform digestive surgery. ^26,27 This system differs in its ability to create three-dimensional imaging as well as having instruments that articulate at the tip. Despite the theoretical advantage of the increased maneuverability of the instruments, this has not translated into shorter operative times, nor has it objectively facilitated the operation. It does, however, add to the complexity of the robotic equipment, necessitating larger operating ports to accommodate the larger instruments. Although the three-dimensional aspect may improve depth of field perception, there are conflicting reports in the literature regarding the advantages of such imaging. ²⁸ To visualize the image in three dimensions, the surgeon must wear specialized binoculars and view the computer screen connected to the robotic computer, thus offering the benefit to the surgeon only, not to the whole surgical team. If, in the future, three-dimensional imaging is deemed beneficial, the Zeus robotic system can easily adapt to such a system. Further experience, including clinical trials and evaluation of training, with the CAS systems are necessary to define the advantages of such additional features and their added cost compared with the objective benefit they bring to the outcome of the surgery.

The current advantage of the routine use of robotics in general surgery is admittedly not readily apparent, because minimally invasive surgery is already the standard for such procedures as cholecystectomies. However, because the surgeon is only one part of the surgical team, we chose to begin clinical trials so that the entire surgical staff, including the operating room staff, anesthesiologists, and surgical assistants, would be able to install and use the equipment and understand the alterations involved in a surgical procedure performed with CAS. Once the safety and feasibility of use of the robotics are shown, as in this preliminary report, future applications and advantages of robotic systems can be explored. One of the potential advantages of the use of the CAS interface is the surgeon’s control of the endoscope by voice command. In our series, a single surgeon was able to coordinate each element of the procedure, eliminating the need for an assistant to manipulate the camera. In other studies, the ability of the computer-controlled endoscope to save positions increased the precision of the manipulations and decreased the time to return to a desired field of vision. ²⁵ Further, the use of the robotics allowed the surgeon to remain in an ergonomically correct position throughout the procedure. This could reduce the fatigue experienced during prolonged or difficult procedures. During our study, a fourth port was used for the placement of clips as well as for a security measure. This port added security because it was accessed at all times by an assistant, who could readily intervene if there were problems with the robotic instruments. However, further experience with CAS and future improvements in the robotic instrumentation, which will include the ability to place clips, should obviate the need for the fourth port. Taken together, the features of the robotic system create a coordinated effort with reduced requirements for assistance during the procedure, making “solo surgery” a possibility. This eventually may greatly reduce the cost of operating room assistance and allow a more coherent procedure.

Another important aspect for the routine use of robotic-assisted surgery is the practicality of its use, including training time and robotic set-up. The experienced laparoscopic surgeons in this study had a period of practice sessions in the laboratory before performing the technique in a clinical setting and therefore were well prepared in the technical aspects of the procedure using the robot. The operating room staff was initially instructed in the set-up of the system by one of the system’s engineers and thereafter installed the system independently, with final positioning performed by the surgeon. The robotic arms were removed from the siderails before the patient was transferred to the gurney for transport to the recovery room. The set-up and removal time required for installation of the robotic arms was approximately 18 minutes, therefore adding only slightly to overall operating room time. The robotic arms are stored on movable chariots that are used for transport to and from the operating room and therefore may be used in any operating suite and do not require special operating room modifications. Overall, these data indicate that the introduction of robotics into the operating room can be achieved with minimal training and robotic support staff and therefore will afford a smooth transition for a laparoscopic surgeon and the operating room staff into the use of the computer interface.

The placement of the ports for the robotic arms presents new challenges and concepts. Each of the robotic arms has a dual sensor system that will disable the robot if any discrepancy is detected in the sensor positions. Lower limits of the robotic arms are set at the time of patient positioning to restrict the movement of the arms before they come in contact with the patient. Therefore, proper positioning of the robotic arms is important so that the procedure can be performed with minimal disruption. In conventional laparoscopy, the ports must be placed to accommodate both the visualization of the field and the position of the surgeon for ease of operating. For robotic-assisted surgery, the ports must be placed in such a manner to minimize robotic arm interference and to maximize visualization of the field. The ports were placed closer to the field than in conventional laparoscopy to aid in visualization. Because the ports are not being directly used by the surgeon, they are not limited by his or her position. In this way, they can be placed closer to the optical port as long as the distance between the operating ports does not cause interference between the robotic arm movements.

Another advantage of the use of the surgical robotics is its unique ability to allow the surgeon to perform surgery at a distance from the patient. The first step toward the realization of this concept occurred with the use of laparoscopic instruments, which allowed the surgeon to perform the operation without being in direct contact with the organs being dissected. The second step is now realized with the use of the robot, where the surgeon is several meters distant from the patient although still in the same operating room. The third application, which is currently in the development stage, will be to have the surgeon located at a remote distance from the patient while performing the operation. Having the surgeon at a distant site would allow for tele-operations and intraoperative expert consultation and intervention, where the surgeon can go to the patient and the on-site surgical team rather than vice versa. This last application will bring new challenges to the surgical community concerning the medicolegal issues of surgery that is a joint effort between different teams. Thus, ethical and medicolegal issues, not the development of the technology itself, may prove to be the limiting factor in the acceptance of these technologic advancements.

Although CAS has led to advances in microsurgery, ergonomics, solo surgery, and telesurgery, its real potential lies in the computer’s ability to integrate information from preoperative imaging studies and planned virtual reality dissection planes during real-time surgery. At present, the surgeon controls the telemanipulated procedure with information projected onto the imaging screen, similar to the conventional laparoscopic technique. However, in the future, this video image may be augmented by projection of anatomical information obtained from computer-generated three-dimensional pre- or intraoperative imaging studies that, in the case of the liver, for example, would define vascular and biliary branches and the exact location of the pathology. In our institute, we have reported on the ability to transform computed tomography scans into digital information that can be rendered by computer manipulation to delineate each patient’s anatomy as well as any underlying pathology. ⁶ This technology will be of the greatest benefit if information from the surgical procedure itself can interact with the anatomical information provided by the preoperative imaging. This link is provided by the use of CAS. With the aid of the computer projecting three-dimensional anatomical reconstructions onto organs throughout the procedure, the surgeon will be able to transform the surgical procedure into an augmented-reality one. In addition, the surgeon will be able to use the three-dimensional imaging reconstruction to simulate the procedure before surgery, using virtual-reality surgery to plan dissection planes and to anticipate difficulties, such as aberrant anatomy. This preoperative simulated procedure has the potential to be programmed into the robotics computer system for guidance during surgery. It is with this aim that we are developing the use of routine robotics in the operating room. Although preliminary, the technology already exists to perform such data integration. It would be an advantage to both patient and surgeon to define the pathology, margins of resection, and vasculature during surgery.

The introduction of CAS into the operating room represents a new era in the history of surgery. As such, the further development and use of CAS must emphasize the dictum of the Hippocratic oath to “first do no harm.” The surgical community must monitor the development and use of this new technology and set standards for its introduction into clinical practice. Surgical robotics is being developed rapidly, and there is an opportunity with this technology to improve surgical techniques greatly. However, its use in the operating room must be carefully considered and evaluated. For this reason, we propose the following guidelines before such advances are implemented in routine practice. First, as with any new technology, experimental laboratory training is necessary before any clinical trials should be considered. The surgeon should be proficient on animal models and should be trained by persons experienced with the use of the robotics in a cooperative effort between industry and surgeons. This should ensure that the surgeon has a high level of expertise using the robotic equipment before using it on a patient. Second, surgical procedures should be performed only within the context of a clinical trial that has a clearly defined protocol that is readily accessible to the patient on request. Third, the risk to the patient should be minimal, and clinical results should approach those of established techniques. If these criteria cannot be met, the advancement of the technology should not take precedence over the patient’s safety.

The current study represents a preliminary effort to establish the safety and feasibility of the introduction of CAS into routine use for deformable organs such as the gallbladder. Throughout the study, we met all the above guidelines. In the future, once the safety and feasibility have been established and confirmed by other studies, CAS can be used to integrate other pertinent computer-generated data, to develop the concept of telerobotic and solo surgery, and to advance complex surgical procedures by the use of the computer to augment the surgeon’s skills. This study represents the first step in a new generation of surgical interventions and advanced procedures, with the ultimate aim of benefiting our patients.