Andrew A. Gumbs, M.D., Nancy J. Hogle, M.S., Dennis L. Fowler, M.D.

Presented as a poster at the Society of Gastrointestinal Endoscopic Surgeons meeting April 29, 2006, Dallas, TX.

Background:
The Global Operative Assessment of Laparoscopic Skills (GOALS) has construct validity in the assessment of surgical residents’ performance of laparoscopic cholecystectomy. We used GOALS in the evaluation of novice and experienced surgical residents to determine whether GOALS would have construct validity in laparoscopic appendectomy.

Methods:
PGY 2 through PGY 5 surgical residents were evaluated by the attending surgeon at the end of a laparoscopic appendectomy. Using the GOALS scale developed by Fried et al., scores for depth perception, bimanual dexterity, efficiency, tissue handling and overall competence were recorded by computer. For analysis, residents were divided into novices (PGY 2 and PGY 3) and experienced (PGY 4 and PGY 5) groups. Biostatistical analysis was performed using single factor ANOVA and the paired T-test.

Results:
25 novices and 18 experienced were included in the analysis. When a mean of all 5 factors was evaluated a statistically significant difference was noted (p=1.97E-06). Difficulty of operation was evaluated in only 30 cases, but no significant differences were noted p=0.21979.

Conclusions:
GOALS is a valid assessment tool for objectively evaluating the technical performance of surgery residents during laparoscopic appendectomy. This study combined with previous validation studies for laparoscopic cholecystectomy document that GOALS is an appropriate assessment tool to evaluate residents’ performance during basic laparoscopic procedures.

Andrew A. Gumbs, M.D., Nancy J. Hogle, M.S., Dennis L. Fowler, M.D.

Presented as a poster at the Society of Gastrointestinal Endoscopic Surgeons meeting April 29, 2006, Dallas, TX.

Presented as a video at the 91st Annual Clinical Congress of American College of Surgeons, October 19, 2005, San Francisco, CA.

Background:
The Global Operative Assessment of Laparoscopic Skills (GOALS), developed by Fried et al., has been shown to have construct validity in the assessment of surgical residents’ laparoscopic skills in dissection of the gallbladder from the liver bed. We hypothesized that GOALS would have construct validity for the entire laparoscopic cholecystectomy procedure.

Methods:
Using GOALS, Attending surgeons evaluated PGY 1 through PGY 5 surgical resident performance during laparoscopic cholecystectomy. Scores for the 5 domains (depth perception, bimanual dexterity, efficiency, tissue handling and overall competence) were recorded by computer. For analysis, residents were divided into 2 groups. The novice group (n=30) included PGY 1 through PGY 3; the experienced group (n=21) included PGY 4 and PGY 5. Biostatistical analysis was performed using single factor ANOVA and the paired T-test.

Results:
When a mean of all 5 factors were evaluated the difference between the groups was (2.88E-06). Difficulty of operation was evaluated in 35 cases; no significant differences were noted (p=0.068).

Conclusions:
GOALS is a valid objective assessment tool for evaluating residents’ performance of the entire laparoscopic cholecystectomy procedure.

Nancy J. Hogle, M.S., Lily Chang, M.D., Mika Sinanan, M.D., Robert Bailey, M.D., and Dennis L. Fowler, M.D.

Presented as a poster at the Society of Gastrointestinal Endoscopic Surgeons meeting April 29, 2006, Dallas, TX.

Presented as an oral presentation at the World Congress of Endoscopic Surgery, Berlin, Germany, September 15, 2006.

Background:
Assessment of operating room performance by videotape review has been attempted without success when using an unvalidated tool. GOALS is a valid assessment tool for objectively evaluating the technical performance of surgery residents during laparoscopic cholecystectomy and laparoscopic appendectomy. We hypothesized that GOALS would reliably differentiate between an experienced (Expert) and an inexperienced (Novice) laparoscopic surgeon have based upon video review of the laparoscopic cholecystectomy procedure.

Methods:
Ten board-certified surgeons actively engaged in the teaching and practice of laparoscopy reviewed the operative performance of one novice and one expert laparoscopic surgeon, based upon a review of the intraoperative videotape recording of the procedure using GOALS. Scores for both the expert and the novice video tape reviews for the 5 domains (depth perception, bimanual dexterity, efficiency, tissue handling and overall competence) were recorded. Biostatistical analysis was performed using single factor ANOVA.

Results:
In the domains of depth perception, bimanual dexterity, efficiency, and overall competence the experienced videotape was scored higher than the novice videotape. There was no difference between the two for tissue handling.

Conclusions:
GOALS is a valid objective assessment tool for evaluating an intraoperative videotape recording of the laparoscopic cholecystectomy.

Nancy J. Hogle, MS, W. Matt Briggs, PhD, and Dennis L. Fowler, MD

Background
The purpose of this study is to document the amount of time required to demonstrate proficiency in two tasks on a virtual reality training simulator in laparoscopic surgery (LapSim®). The modeling of the general learning curve is useful in designing training programs.

Methods
Twenty-nine medical students participated in eight attempts of two tasks: camera navigation (CN), a simple task and lifting and grasping (LG), a complex task. For each attempt at CN the following variables were recorded by computer: time to completion, drift, number of misses, path length, angular path length, number of instances of tissue damage, and maximum tissue damage. For each attempt at LG, time to completion, number of misses for the left (L) and right (R) hand, L and R path length, L and R angular path length, number of instances of tissue damage, and maximum depth of penetration of tissue damage were recorded. Learning curves for all 16 outcome variables of the CN and LG tasks were generated after eight attempts.

Results
A discernable amount of learning occurred for most subjects as evidenced by a decrease in value of all variables. This decrease for each person is called a learning curve. A plateau in the learning curve (learning has occurred) was achieved within 8 attempts for 3CN and 1 LG variables. A plateau in the learning curve was nearly achieved for 4 CN and 2 LG variables. Six LG variables had a downward trend to the mean learning curve over eight sessions, but did not reach a plateau.

Conclusion
A learning curve for each outcome variable for CN and LG has been documented for rank novices. For many of the variables, the learning curve reached a plateau in 8 hours or less, indicating that 8 attempts is enough for the purposes of training. For some variables, the learning curve did not reach a plateau, which indicates the task is more complex and that more training is required.