Skill acquisition process in vascular anastomosis procedures: a simulation-based study.

Objectives: There is growing evidence that practice on simulation models can improve technical skills in surgery. The aim of this study is to assess the effects of our tissue-based simulation model of vascular anastomosis on skill acquisition.

Methods: Five junior (Group I) and five senior (Group III) cardiovascular surgery residents, and five surgeons from different surgical departments (Group II) attended the study. A total of 180 vascular anastomoses on a bovine heart simulation model were performed in a 3-month period; each group performed 20 anastomoses per month (each participant in each group conducted four anastomoses per month). The anastomoses were evaluated according to criteria including, duration of the procedure, existence of anastomotic leak, additional suture requirements, matching between graft diameter and arteriotomy length, patency rates and inadvertent posterior wall injuries. Each practice was recorded with a video camera and eventually reviewed by three cardiovascular surgeons, who were blinded to groups. Results were compared for analysing the skill acquisition process in each group.

Results: The mean anastomosis time (Group I: 22.25 ± 2.02, 18.10 ± 0.78, 15.00; Group II: 17.05 ± 1.39, 15.45 ± 0.82, 13.00 ± 0.79; Group III: 13.65 ± 0.67, 11.45 ± 1.14, 10.50 ± 1.10) and additional suture requirements (Group I: 1.95 ± 0.68, 1.30 ± 0.80, 1.00 ± 0.32; Group II: 1.80 ± 0.41, 1.45 ± 0.60, 1.45 ± 0.60; Group III: 0.65 ± 0.48, 0.40 ± 0.50, 0.40 ± 0.50) decreased gradually (P < 0.0001 for each) in all groups. There was statistically significant improvement over time in anastomotic leakage (Group I: 90, 65, 20%; Group II: 50, 25, 5%; Group III: 20, 25, 5%), match between the arteriotomy and the graft (Group I: 35, 25, 75%; Group II: 60, 45, 85%; Group III: 85, 65, 95%), posterior wall injury (Group I: 70, 50, 15%; Group II: 50, 30, 5%; Group III: 30, 30, 5%) and patency (Group I: 45, 15, 75%; Group II: 60, 50, 95%; Group III: 80, 85, 95%) in all groups, except for the occurrence of anastomotic leaks and patency rates in the senior cardiovascular resident group (Group III).

Conclusions: Although the most significant improvement was observed in Group I, all groups demonstrated improved skills with the simulation model. Therefore, it can be suggested that anastomosis training on tissue-based simulation models may be beneficial for the skill acquisition process.