Validating novel portable three-dimensional models based on MISTELS for training and evaluation of laparoscopic skills. - Are they good enough to prepare surgeons for 2D and 3D based laparoscopic surgery?

Validating novel portable three-dimensional models based on MISTELS for training and evaluation of laparoscopic skills. – Are they good enough to prepare surgeons for 2D and 3D based laparoscopic surgery?

Alpa J Morawala, MBBS, MSc¹, Walid S Elbakbak, MBBCh, MSc, MRCS¹, Badriya Alaraimi, MD, MSc, MRCS², Bijendra Patel, FRCS². ¹Queen Mary University of London, Bart Cancer Institute, ²Royal London Hospital

Introduction:This study is commenced to establish validity of newly developed 3D models using (MISTELS) criteria for objective assessments and training of laparoscopic skills in 3D, and to compare proficiency criteria with 2D.

Materials and Methods:Laparoscopic experts involved in designing 3D models. Models were built to use inside a training box. Total 21 subjects [10 novices, 7 intermediates, 4 experts] followed the training flow comprises of three sets of repetitions of an improved FLS tasks (Peg transfers, Ligating Endoloop, Intracorporeal suture, Pattern-cutting) and an additional new task (Creating Zig-zag loop). Novices initiated study with 3D followed by 2D, whereas other two groups started with 2D then 3D. Total scores and total errors were measured. Total score calculated by subtracting penalty score from performance time. Lower score reflects better performance. Student’s t test used to compare the data (mean value of total score) of each group.

Results: Longer performance time with higher errors (Table 2) noticed in novices and intermediates in both visions. Fewer errors (Table 2) and less time observed in experts. Each group performed significantly better in 3D. In addition, a group started the study in 3D showed an improvement when asked to perform in 2D. Statistically significant differences (Table 1) found between skilled (Expert) and non-skilled group (Novices/ Intermediates). Using Likert scale (1-5); candidates rated models (face validity); 3.3 in 2D and 4.3 in 3D (better depth perception).

Table 1:P value for Total score
	Expert Vs. Intermediate	Expert Vs. Novice
2D	0.02	0.00
3D	0.08	0.00

P value for total error
	Expert Vs. Intermediate	Expert Vs. Novice
2D	0.04	0.03
3D	0.18	0.39

Table 2: Total Error Mean value (task 1-5)
GROUPS	Total errors (mean) 2D	Total errors (mean)3D
NOVICES	2.2	1.9
INTERMEDIATES	1.9	0.7
EXPERTS	1.1	0.5

Conclusions: We have effectively established the construct validity of novel models for both visual modalities. Hence, models can be successfully used for simulation training in both visual modalities. 3D vision allows remarkable improvement in performance and error rates compared to the 2D vision. Importantly, advanced laparoscopic skills learned with 3D models allow transferrable skills from 3D to 2D.