Noninvasive Brain Imaging Demonstrates that Surgical Skills Transfer from Training Simulators to Ex Vivo Models

e22

Scientific Forum: 2017 Clinical Congress

CONCLUSIONS: There is a significant increase in SSI and other perioperative complications in patients with a BMI > 35 kg/m2 undergoing AWR. Higher BMI is also independently associated with higher resource utilization in this patient population. Severely obese patients in need of AWR may benefit from a structured preoperative weight loss intervention. Multiscale Modeling of Surgical Flow Explains Lognormal Distribution of Surgical Time and Occurrence of Large Delays Marc Garbey, PhD, Juliette Rambourg, Guillaume Joerger, Brian J Dunkin, MD, Barbara L Bass, MD, FACS Houston Methodist Hospital Research Institute, Houston, TX INTRODUCTION: Large hospitals’ challenge is to quantify, evaluate, and improve the efficiency of surgical suites. Its complex environment involves hundreds of staff and patients per day. We propose to supplement the surgical suites management system and bring effective human-friendly computerization for staff coordination. Teamwork quality profoundly impacts the optimum management of resources. We model that complexity with a system approach to get a better understanding of the root cause analysis of delay and poor efficiency. METHODS: Our mathematical model of surgical flow is built upon observations and robust clinical data with a noninvasive array of sensors that automatically monitors the surgical flow. The data covers operating room (OR) status for a thousand procedures. Staff rescheduling decisions and coordination were observed and analyzed with a campaign of photography of the collaborative case scheduling whiteboard for weeks. Comparisons between OR status and this whiteboard were performed on about 40 cases. RESULTS: The model reproduces the statistic distribution pattern over 1 year of performance indicators: turnover time, induction of anesthesia time, time between extubation and OR exit. The model classifies human factors’ impact and limitation of shared resources on flow efficiency. We found that communication delays and suboptimal OR awareness in large surgical suites have significant impacts on performance and should be addressed. CONCLUSIONS: The future of complex surgical flow management may rely on cyber-physical infrastructures that use real time simulation with multiscale model like ours to assist the management process and fully utilize the surgical suite resources without overloading the staff.

Noninvasive Brain Imaging Demonstrates that Surgical Skills Transfer from Training Simulators to Ex Vivo Models Arun Nemani, PhD, Clairice Cooper, MD, Xavier Intes, PhD, Suvranu De, PhD, Steven D Schwaitzberg, MD, FACS Rensselaer Polytechnic Institute, Troy, NY; University of Buffalo, Buffalo, NY INTRODUCTION: Published research shows that surgical simulation trainers provide effective and safe ways to acquire technical

J Am Coll Surg

surgical motor skills; however, few published studies conclude that these motor skills transfer from simulation to operative environments. The purpose of this study is to measure surgical motor skill transfer from training simulators to ex vivo tissue models using brain activation and task performance metrics. METHODS: Of 19 participating medical students, 6 practiced more than 100 physical FLS pattern cutting tasks, 8 practiced over 100 virtual FLS pattern cutting tasks, and 5 control students had no practice. Students performed the tasks on ex vivo cadaver peritoneum tissue mimicking the simulation environments. Performance scores were calculated using official FLS metrics and transfer task scores were quantified by total completion time. Since the primary motor cortex (M1) is heavily involved in the stages of motor skill learning, brain activation was quantified by the change in oxygenated hemoglobin concentration in M1 measured by functional near-infrared spectroscopy. RESULTS: Results indicate that both the physical trainer (7.93.3 min) and virtual trainer (13.052.6 min) groups completed the transfer task significantly faster than the control group (15.55.6 min, p<0.05). Furthermore, cortical activation was significantly higher for the FLS group in the M1 compared to the control group (p<0.05). CONCLUSIONS: Motor skills transfer from simulation to ex vivo environments based on brain activation and task performance metrics. This type of measurement will provide valuable insight into competency based skill assessment as well as skill degradation and recovery assessment. Outcome of Transvaginal Natural Orifice Transluminal Endoscopic Surgery (NOTES) Cholecystectomy: Data from the German NOTES Registry Kai S Lehmann, MD, Carsten Klinger, Dirk R Bulian, Jens Burghardt, Carsten Zornig, Heinz J Buhr, MD Charite, Campus Benjamin Franklin, German Society for General and Visceral Surgery, Berlin, Germany INTRODUCTION: The German NOTES (GNR) registry started in 2008 and currently is the largest NOTES outcome database. We present the outcome data for transvaginal hybrid cholecystectomies (tvChole, StuDoQ NOTES registry, German Society for General and Visceral Surgery). METHODS: The GNR records demographical and therapy data as well as data on the postoperative course in an online database. All patients with tvChole between March 2008 and February 2017 were included in the analysis. Data are presented as mean  standard deviation or percentages (counts). Multivariate regression was performed for age, BMI, cholecystitis, and institutional case volume. RESULTS: TvCCE were the most frequent operation with 3,694 (88%) of 4,198 valid cases in the GNR. All tvChole were performed in hybrid technique with transvaginal access and one or more percutaneous trocars (1.20.5). Mean institutional case