Development and validation of a porcine organ model for training in essential laparoscopic surgical skills

Abstract

Objectives: To develop a wet laboratory training model for learning core laparoscopic surgical skills and evaluating learners' competency level outside the operating room.

Methods: Participants completed three tasks (task 1: tissue dissection around the aorta; task 2: tissue dissection and division of the renal artery; task 3: renal parenchymal closure). Each performance was video recorded and subsequently evaluated by two experts, according to the Global Operative Assessment of Laparoscopic Skills and task-specific metrics that we developed (Assessment Sheet of Laparoscopic Skills in Wet Lab score). Mean scores were used for analyses. The subjective mental workload was also assessed (NASA Task Load Index).

Results: The 54 participants included 32 urologists, eight young trainees and 14 medical students. A total of 13 participants were categorized as experts (≥50 laparoscopic surgeries), eight as intermediates (10-49) and 33 as novices (0-9). There were significant differences in the Global Operative Assessment of Laparoscopic Skills and Assessment Sheet of Laparoscopic Skills in Wet Lab scores among the three groups in all three tasks. Higher NASA Task Load Index scores were observed in novices, and there were significant differences in tasks 1 (Kruskal-Wallis test, P = 0.0004) and 2 (P = 0.0002), and marginal differences in task 3 (P = 0.0745) among the three groups.

Conclusions: Our training model has good construct validity, and differences in the NASA Task Load Index score reflect previous laparoscopic surgical experiences. Our findings show the ability to assess both laparoscopic surgical skills and mental workloads, which could help educators comprehend trainees' level outside the operating room. Given the decreasing opportunity to carry out pure laparoscopic surgeries because of the dissemination of robotic surgery, especially in urology, our model can offer practical training opportunities.

Keywords: animal organs; laparoscopic surgery; simulation; surgical education; wet lab training.

Fig. 1

Photographs of the simulation training. (a) Training view. Candidates were informed of training slots, prepared on the Doodle website (calendar tool for time management and coordinating meetings), via email. Thereafter, participants voluntarily booked convenient slots according to their schedules, and participated in the training. (b,c) Task 1 (tissue dissection around aorta). (d) Task 2 (tissue dissection and divide renal artery). In tasks 1 and 2, laparoscopic scissors (Scissors Metzenbaum; Olympus, Tokyo, Japan), laparoscopic grasping forceps (CLICKline CROCE‐OLMI Grasping Forceps; Karl Storz, Tokyo, Japan) and a laparoscopic clip applier (Hem‐o‐lok Endoscopic Appliers Large; Teleflex, Tokyo, Japan) were used. (e) Task 3 (renal parenchymal closure). In task 3, laparoscopic needle holders were used (KOH Macro Needle Holder, ratchet position right, jaws curved to left, and KOH Macro Needle Holder, ratchet position left, jaws curved to right; Karl Storz). (f) Box trainer. (g) Setting of aorta in task 1. (h) Setting of kidney in task 2. (i) Setting of kidney in task 3.

Fig. 2

GOALS scores at the time of participants’ first training session divided by previous experience of laparoscopic surgery. (a) Task 1. (b) Task 2. (c) Task 3. There were significant differences in GOALS scores among the three groups in all three tasks.

Fig. 3

ALL scores at the time of participants’ first training session divided by previous experience of laparoscopic surgery. (a) Task 1. (b) Task 2. (c) Task 3. There were significant differences in ALL scores among the three groups in all three tasks.

Fig. 4

NASA‐TLX scores at the time of participants’ first training session divided by previous experience of laparoscopic surgery. (a) Task 1. (b) Task 2. (c) Task 3. Higher NASA‐TLX scores were observed in novices, and there were significant differences in tasks 1 (P = 0.0004) and 2 (P = 0.0002), and marginal differences in task 3 (P = 0.0745) among the three groups.

Fig. 5

ROC curves of tasks 1, 2 and 3 for classifying the ESSQ qualification status based on GOALS score. All three tasks showed good separability.

Fig. 6

Tissue similarity and effectiveness of each task evaluated by the experts and intermediates. Most of the aspects were rated as above average, except for fat tissue. Both the intermediates and experts rated all three tasks as being effective for training.

Fig. 7

Learning curves of task 1 in the 15 participants who underwent the training multiple times. (a) GOALS score. (b) ALL score. (c) NASA‐TLX score. No constant trend was observed among the participants, although an increasing tendency in GOALS and ALL scores, and a decreasing tendency in NASA‐TLX scores were observed in several participants.