512 The Learning Curve for Transvaginal Hybrid Notes Cholecystectomy

The Learning Curve for Transvaginal Hybrid Notes Cholecystectomy Kai S. Lehmann*, Carsten Zornig, Michael Butters, Thomas Kocher, Jens Burghardt, Georg Arlt, Norbert S. Runkel, Dirk R. Bulian, Konstantinos Damanakis, Jorg Koninger, Jorgen Bretschneider, Heinz J. Buhr Department of Surgery, Charite University Hospital—Campus Benjamin Franklin, Berlin, Germany; Department of Surgery, Israelitisches Krankenhaus, Hamburg, Germany; Department of Surgery, Krankenhaus Bietigheim, Bietigheim-Bissingen, Germany; Department of Surgery, Kantonsspital Baden, Baden, Switzerland; Department of Surgery, Evangelisch-Freikirchliches Krankenhaus Rudersdorf, Rudersdorf, Germany; Department of Surgery, ParkKlinik Weissensee, Berlin, Germany; Department of Surgery, Schwarzwald-Baar-Klinikum, Villingen-Schwenningen, Germany; Department of Surgery, Krankenhaus Merheim, Koln, Germany; Department of Surgery, Petrus Krankenhaus, Wuppertal, Germany; Department of Surgery, Katharinenhospital, Stuttgart, Germany; Department of Surgery, Asklepios Klinik Weissenfels, Weissenfels, Germany Background: Learning curves for Natural Orifice Transluminal Endoscopic Surgery (NOTES) have not been established so far. The German NOTES Registry (GNR) currently is the largest NOTES outcome database worldwide. 89% of all procedures in the GNR are transvaginal hybrid NOTES cholecystectomies (tvChole). The aim of this study was to examine the learning curve for tvChole. Methods: The GNR records NOTES procedures in German-speaking countries since March 2008. All tvChole in the GNR have been analyzed for learning effects. The main outcome parameters were length of surgery, conversions and complications. High volume centers ( 100 procedures) were compared with low volume centers ( 100 procedures). A multivariate regression analysis was performed for institutional case volume (cumulative case volume of individual centers), patient age and patient BMI. Results: 1823 tvChole were recorded in the GNR. Length of surgery showed a distinct learning curve which correlated to the institutional record number (procedure no. of the center at the time of surgery, figure 1). High volume centers operated faster (51.6 21.8 min) than low volume centers (72.0 27.8, P 0.001). However, the complication rate did not differ significantly between these two groups (4.3% vs. 5.1%, P 0.397). Age and BMI (positively correlated), as well as institutional case volume (negatively correlated) strongly influenced length of surgery and length of hospital stay in the multivariate analysis (table 1). Institutional case volume had a small effect on the conversion rate (P 0.041), but no effect on intraoperative (P 0.100) or postoperative complications (P 0.395). Discussion: Learning effects for length of surgery and hospital stay were observed. The learning curve for length of surgery reaches a plateau after 50 procedures. Institutional case volume as a measure of experience had a small effect on the conversion rate, but no effect on complications. Possible reasons are a selected patient group and highly experienced surgeons, as tvChole is usually performed only by one or two senior surgeons in each participating center. Conclusion: Transvaginal hybrid NOTES cholecystectomy has a learning curve, but it is safe even in low volume centers when performed by experienced surgeons.