AISP - 36^th National Congress. Bologna, Italy. October 4-6, 2012

Laparoscopic Left Pancreatectomy: Does Exist a Learning Curve?

Claudio Ricci, Salvatore Buscemi, Marielda D’Ambra, Giovanni Taffurelli, Carlo Alberto Pacilio, Eugenia Peri, Raffaele Pezzilli, Nicola Marrano, Riccardo Casadei, Francesco Minni

Department of Surgery, “S.Orsola-Malpighi” Hospital. Bologna, Italy

Context Surgeons are performing laparoscopic left pancreatectomy (LLP) with increasing frequency. To our knowledge do not exist studies which defines learning curve (LC). Objectives To define a number of procedures needed to achieve LC in LLP. Methods From January 2008 to June 2012 data regarding 25 patients, undergoing LLP for pancreatic lesions, were collected in a prospective database. All procedures were performed by a single high volume surgeon in advanced laparoscopic and open pancreatic surgery. Decrease of the operative time (OT) was used as parameter to establish the achievement of LC. A preliminary multivariate analysis was carried out to establish which factors influenced OT. Correlation between OT and cumulative sum of procedure (CUSUM) was evaluated to calculate the LC cut-off. Finally multivariate analysis was repeated including LC cut-off. ANOVA test was used to estimate correlation and to calculate multivariate model. Results There were 18 (72%) females and 7 (28%) males with mean age of 55±16 years and mean BMI of 27±5 kg/m². Patients were more frequently ASA II (64%). Thirteen patients (52%) presented one or more co-morbidity and had a previous surgical abdominal procedure. Splenectomy was carried out in 18 (72% ) cases. An extended resection was conducted in 5 cases (20%). Mean operative time was 219±52 min. Rate of conversion was 16%. Pathological examination showed only in 11 (44%) cases a malignant disease and none ductal adenocarcinoma. Preliminary multivariate analysis showed that splenectomy significantly decreased OT while size of lesion increased OT (P=0.033 and P=0.041, respectively). A significant inverse correlation was found between OT and CUSUM (P=0.050) and the LC cut-off was 14 procedures. Final analysis including LC cut-off showed that the achievement of LC cut-off reduced significantly OT (P=0.047). Instead BMI and extended resection independently increased OT (P=0.030 and P=0.022, respectively) Conclusion In our experience the number of procedure needed to achieve LC was 14 LLP. BMI and extended resection influenced OT even after the achievement of LC.