부산대학교 도서관

Background Interest in natural orifice transluminal endoscopic surgery (NOTES) has expanded, and the first experiences with patients using different techniques have been reported. However, no work has addressed the learning process or the limitations of the procedures. The relation between inexperience and complications became a major concern after the introduction of laparoscopic surgery. This study investigates the learning process for a new technique using specially designed instruments in an ex vivo model before clinical application. Methods Specially designed instruments and a single-port technique using the Tuebingen Trainer were used to evaluate instrument and surgeon performance (learning curve) in terms of time and errors. A total of 90 procedures performed by three surgeons were evaluated. Group and individual learning curves were plotted. Results All the surgeons showed a reduction in both mean cholecystectomy time (subject A: 27.2 vs 16.6 min subject B: 21.4 vs 19.22 min subject C: 21 vs 19.7 min) and mean errors (subject A: 2.8 vs 1.6 subject B: 3.5 vs 2.6 subject C: 3.5 vs 2). A plateau was reached after approximately 15 procedures. Group learning curve analysis showed a significant reduction in time between the first group (mean, 24.97 +- 5.8 min) and last group (mean, 19.30 +- 3.09 min F[1,28] = 11.83 p = 0.001) for 15 procedures, as well as reduced technical errors in the fifth group, from 3.7 +- 1.65 to 1.6 +- 1.04 (F[1,28] = 8.90 p < 0.01), demonstrating a learning effect. The number of optic and access port position changes were recorded, setting a standard for normal instrument performance. Conclusion This study shows that the tasks of cholecystectomy can be learned safely in a reasonable number of simulations with the new instruments. Although this is a new technique, prior laparoscopic surgery experience is helpful. The technique offers an advantage over those using flexible endoscopes.