• conventional laparoscopy
• laparoendoscopic single-site surgery
• sensor-based laparoscopic simulator
• skills evaluation
• tissue manipulation force
• trocar sensor

Background Since the introduction of laparoscopic surgery, experts have been interested in its effectiveness, virtues and drawbacks. The amount of operations performed using conventional laparoscopy and laparoendoscopic single-site surgery has progressively increased over the past years. Both approaches to laparoscopy were shown to be very effective, nevertheless, limitations still persist. Despite the increased number of clinical applications of single-site approaches to laparoscopic surgery, limited research in a simulated environment has been done to evaluate its performance and performants. The impact of surgical consequences led to the development of laparoscopic simulators with the goal of training novice laparoscopic surgeons. Developing technical skills is essential to surgical training and thus the objective of the present thesis is to compare the performance and learning progress of experienced surgeons and students of conventional and single-site laparoscopy.

Physical Setup For the purpose of the present study, a laparoscopic physical simulator was developed from a basic shell. Participants interact with the simulator through conventional straight or double curved 5 mm hand instruments. Different configurations of the hand instruments with 5mm trocars or a single-incision port were analysed. A force platform, inserted inside the simulator, measured parameters related with tissue manipulation, including different types of forces applied. Furthermore, handle sensors that measure grasp forces between the pincers of the conventional hand instruments were developed. A trocar sensor was built to obtain data about the forces transferred from the hand instruments to the abdominal wall. On the top of the force platform, a custom-made training platform was placed, on which participants had to perform a task. This manipulation task and platform were developed for the purposes of this thesis with the goal of mimicking some basic actions performed during laparoscopic surgery. During the task, a silicone piece of artificial tissue had to be navigated through a ring and attached to a pin.

Experimental Setup Twenty-four medical students without surgical experience were randomly assigned into two groups. The first group performed trials on the simulator using the conventional laparoscopic approach (hereby abbreviated as Conv) and later switched into single-site laparoscopic approach with conventional straight instruments (from now on abreviated as LESS (C)). The second group performed the same training in reverse order. Each approach was executed six times by each group, totalling 288 trials. The performance of the participants on the tissue manipulation task was assessed. For comparison, a control group of five single-site laparoscopy expert surgeons were asked to execute the same task as the two groups of novices. In random order, three trials were performed with Conv. approach, three trials with LESS (C). Additionally, experts performed three trials with single-site approach with double-curved hand instruments (hereby LESS (D)), not performed by novices.

Experimental Study Two main studies were carried out. The first study compares tissue manipulation performance of novices and surgeons. The second study concerns only the control group of expert surgeons. The latter focus on the amount of grasp force applied on the tissues and the amount of forces applied on the trocar (forces that in a real operation would be applied to the incision on the abdominal wall). After the conclusion of the trials, novices and experts filled in a questionnaire that evaluated their performance. This questionnaire was used to understand the connection between self-evaluation and quantitative results from the simulator.

First Study Results Time, maximum applied force (pushing and pulling) and force-volume (a parameter based on the variation of the force applied) are fundamental for the determination of the performance level of the participants. As a summary of the results, the conclusion reached was that the duration of a single trial, the amount of maximum interaction force and force-volume applied improve as the task is repeated. This is a positive result as it demonstrates that these are valid parameters to assess the skill of a surgeon or medical student.

Experts demonstrated their proficiency with a short duration of the trials (48s std 20.8 for Conv; 60s std 29.9 for LESS (C); 88s std 26.3 for LESS (D)). In comparison, novices performance on their final trial (group 1: 101s std 68.1 for Conv; 193s std 151.4 for LESS (C); group 2: 115s std 40.4 for Conv; 199s std 99.1 for LESS (C)) did not reach the same proficiency as experts. Novices exhibited a progressive improvement in the time taken to perform the task from trial to trial.

Another important parameter that evidenced progressive improvement during the conventional approach, was the maximum interaction force. Nevertheless, the maximum interaction force during LESS (C) approach was characterised by fluctuating values in the first group and increasing values in the second group. The results obtained from the first group of novices (6.6N std 2.3 for Conv and 9.0N std 2.7 for LESS (C)) were similar to those of experts (7.5N std 3.2 for Conv and 9.0N std 2.2 for LESS (C)). Slightly higher results, in what concerns maximum force applied, were obtained in group 2 (8.0N std 2.2 for Conv and 10.9N std 2.5 for LESS (C)). Data about maximum interaction force according to the LESS (D) trials was obtained only from the control group (8.9N std 2.6).

The third important parameter for the evaluation of participants performance was the force-volume. Force-volume is calculated using the standard deviations of the exerted forces in a 3-dimensional referential. This parameter is related to the smoothness of a movement and dexterity of using the laparoscopic hand instruments. If the silicone artificial tissue is pushed with a constant force in one direction, a small force-volume is obtained. During the conventional approach, the first group of novices presented the lowest value of force-volume (4.9N^3 std 6.2) in comparison with the second group (9.4N^3 std 7.1) and almost twice lower than the experts (9.6N^3 std 13.1). The force-volume during LESS (C) approach, exerted by the second group of novices (11.8N^3 std 15.4), presented the lowest value, in this approach, among the novices (18.8N^3 std 17.9) and even among the experts (30.2N^3 std 53.8). As novices performed a larger number of trials, the fact that novices used less force-volume than the experts was interpreted as tangible proof of evidence to the importance of practice.

Second Study Results Grasp forces (between the pincers) applied by experts were obtained and analysed. The mean grasp force applied by the left (non-dominant) hand (8.4N std 2.7 during Conv and 9.5N std 2.9 during LESS (C)) was weaker in comparison with the grasp forces by the right hand (7.7N std 4.7 during Conv and 8.7N std 4.1 during LESS (C)). No significant difference in the maximum grasp force was observed between the left hand (19.6N std 1.1 during Conv and 19.4N std 1.1 during LESS (C)) and the right hand (18.0N std 6.0 for Conv and 19.7N std 3.6 for LESS (C)).

The mean force detected by the trocar sensor did not show excessive values (2.7N std 0.5 for LESS (C) and 2.7N std 0.3 for LESS (D)). The maximum force detected by the trocar sensor (11.4N std 3.7 for LESS (C) and 9.3N std 2.9 for LESS (D)) requires definitely further investigation as the value which causes damage in patients is not known.

Additional Results The questionnaire showed that the first group of novices preferred the conventional approach, while only 36% of the participants from the second group expressed preference about LESS (C) approach. This difference in preference can be justified by the order in which the approaches were learnt. In general, there was a consensus about the self-evaluation of the participants performance. Both groups of novices rated the conventional approach with 8 out of 10, while the LESS (C) approach was rated with 6 in group 1 and 7 in group 2.

Further results were obtained from direct observation of tissue handling and posture of the participants. Different handling strategies of the artificial tissue, as well as the correct posture of the participants influenced the overall performance.

Conclusion The sensor-based laparoscopic simulator built for the present study, can be considered a reliable, valid and innovative device for acquiring skills in the performance of laparoscopic tasks. Useful results about interaction forces, grasp forces, force-volume and trocar forces were obtained. The order of the approaches, the handling strategies and the correct posture of the participants, showed to influence the overall performance on the laparoscopic simulator.

Further research is required to determine a reference threshold that indicates a limit to the range of safe forces applied by novice surgeons. This would allow the use of the simulator as a training device that gives feedback on the performance of the trainee. Additionally, it is relevant to understand if strong forces, applied to the incision of the abdominal wall, during manoeuvring of the hand instruments are a significant reason for post-operative complications.