Oral (pre-recorded)
Pancreas 		  
OP04 Pancreas: Surgical Outcomes 
Selection of Presentations from Abstract Submissions
OP04-02 Laparoscopic Distal Pancreatectomy Shortens Hospital Stay: Results from a Single-Center, Randomized Controlled Trial (LAPOP)
Bergthor Björnsson, Sweden

B. Björnsson1, A. Lindhoff Larsson1, C. Hjalmarsson2,3, T. Gasslander1, P. Sandström1
1Department of Surgery, University of Linköping, Sweden, 2Department of Surgery, Blekinge Hospital, Sweden, 3Department of Clinical Sciences, Lund University, Sweden

Introduction: Nonrandomized retrospective studies have suggested that laparoscopic distal pancreatectomy (LDP) is advantageous compared to open (ODP) regarding hospital stay, blood loss and recovery. The only randomized study is available shows enhanced functional recovery after LDP.
Methods: Sixty patients, evaluated at a multidisciplinary tumor board and planed for standard distal pancreatectomy were prospectively randomized to LDP or ODP in a parallel group, single-center superiority trial. The primary outcome was postoperative hospital stay with the hypothesis that LDP would shorten it.
Results: Fifty-eight patients, 34 male and 24 female, were assigned to LDP (n=29, mean age 68 years) vs. to ODP (n=29, mean age 63 years) and included in a intention-to-treat analysis. The postoperative hospital stay was 5 (IQR 4-5) days in the LDP group vs. 6 (5-7) days in the ODP group (P=0.002). Functional recovery was reached after 4 (2-6) vs. 6 (4-7) days (P=0.007), and the operation time was 120 minutes in both groups (P=0.48). Blood loss was reduced with LDP,
50 (25-150) compared to 100 mL (100-300) (P=0.018). No difference was found in the complication rates with 4 vs. 8 patients in the LDP and ODP groups respectively experiencing complication of Clavien-Dindo grade 3 or higher. Similarly, the rate of post pancreatectomy fistula did not differ between the groups (9 vs. 11 patients).
Conclusions: LDP is associated with shorter hospital stay, enhanced functional recovery and less bleeding as compared to ODP, and should therefore be considered the as the treatment standard for patients in need of distal pancreatectomy.
OP04-03 Factors Associated with Overall Survival in Pancreatic Cancer Treated with Neoadjuvant Therapy and Surgery
Rebecca Kim, United States

R. Kim1, S. Tsai1, M. Aldakkak1, B. George2, M. Kamgar2, B. Erickson3, N. Kulkarni4, D. Evans1, K. Christians1
1Surgery, Medical College of Wisconsin, Milwaukee, United States, 2Medical Oncology, Medical College of Wisconsin, Milwaukee, United States, 3Radiation Oncology, Medical College of Wisconsin, Milwaukee, United States, 4Gastroenterology, Medical College of Wisconsin, Milwaukee, United States

Introduction: Median survival following surgical resection of pancreatic cancer (PC) has increased due to advances in chemotherapy, radiotherapy, and surgical technique. We examined clinical factors associated with overall survival (OS) in patients with PC who received neoadjuvant therapy and surgery.
Method: We conducted a retrospective review of a prospectively maintained PC database at high-volume referral center. Patients with non-metastatic PC who received neoadjuvant therapy and underwent surgical resection between 2009 and 2019 were included. Demographic, clinical, and pathologic variables were examined with Cox proportional hazards models to identify prognostic factors on OS.
Results: Neoadjuvant therapy and surgery was completed in 460 consecutive patients; 227 (49.4%) patients were female. The median age was 64 (IQR[58, 71]) years. Median OS was 40.2 (IQR[23.4, 87.3]) months. On bivariate analysis, elevated CA19-9 prior to surgery
(HR 1.8 [1.4,2.3]), vein resection (HR 1.6 [1.2, 2.1]), lymphovascular invasion (HR 1.5 [1.1, 2.0]), positive superior mesenteric artery margin (HR 1.8 [1.2, 2.7]) and higher nodal stage were associated with worse OS. On multivariate analysis, shorter OS was associated with increased CA19-9 (HR 1.6 [1.2,2.2]), higher nodal stage (HR 1.5 [1.0, 2.1] for N1; HR 1.9 [1.2, 3.0] for N2), and liver as the first site of disease recurrence. Preoperative clinical stage was not a significance predictor of OS.
Conclusions: We report OS outcomes from one of the largest cohorts of resected PC in the era of neoadjuvant therapy. Preoperative CA19-9 after neoadjuvant therapy was associated with OS, emphasizing the impact of treatment response on OS.
OP04-04 Recurrence Following Neoadjuvant Therapy and Resection for Pancreatic Ductal Adenocarcinoma: A Comprehensive Meta-Analysis and Meta-Regression
Bathiya Ratnayake, New Zealand

B. Ratnayake1, A. Savastyuk1, J. Hammond2, G. Sen2, S. White2, J. French2, S. Pandanaboyana2
1Department of Surgery, University of Auckland, New Zealand, 2Department of Hepatobiliary, Pancreatic and Transplant Surgery, Freeman Hospital, United Kingdom

Background: This review aims to provide a comprehensive analysis of recurrence patterns in patients undergoing neoadjuvant therapy (NAT) in comparison to those undergoing up-front surgery (US) for PDAC.
Methods: The EMBASE, SCOPUS, PubMed and Cochrane library databases were systematically searched to identify eligible comparative studies. The primary outcome was time to first recurrence and location of recurrence.
Results: Twenty-five articles were identified including 4822 patients undergoing resection. The weighted mean follow-up interval for recurrence outcomes was 40.8months (CI 33.4-48.1). The weighted mean overall recurrence rate was 63.4% (CI 51.8-73.%) for NAT, significantly lower than the 74% (CI 68.7-80%) weighted overall recurrence rate of the US cohort (OR 0.67 (CI 0.52-0.87), P=0.006). NAT was also associated with a significantly longer weighted mean time to first recurrence (NAT 18.8months US 15.7months, P=0.015). The weighted locoregional recurrence (NAT 12%, US 27%, P=0.004) and liver recurrence (NAT 19.4%, US 30.1% P=0.023) rates were markedly improved among NAT patients. Weighted lung and peritoneal recurrence rates did not differ (P=0.705 and P=0.549 respectively). NAT was further associated with a greater two- (NAT 39%, US 22% OR 1.84 (CI 1.22-2.78), P=0.007) and five-year (NAT 24%, US 13% OR 1.95
(CI 1.03-3.69), P=0.043) recurrence free survival. Borderline resectability, presence of perineural invasion and a lower N0 nodal status were positive predictors of overall recurrence in the NAT cohort.
Conclusions: NAT is associated with improved overall recurrence rates and longer time to first recurrence, an observation likely attributed to the improved rates of locoregional and liver recurrence.
OP04-05 Efficacy of Peri Operative Hydrocortisone and Indomethacin Treatment in Reducing Major Complications after Whipple's Pancreaticoduodenectomy, Randomized Controlled Clinical Trial
Rohit Dama, India

R. Dama1, K. Kant1, P. Rebala1, G. Rao1, D. Reddy2
1Surgical Gastroenterology, Asian Institute of Gastroenterology, India, 2Medical Gastroenterology, Asian Institute of Gastroenterology, India

Introduction: Post Whipple's Pancreatico-duodenectomy (PD) the major concern is pancreaticojejunostomy (PJ) leak which leads to post operative pancreatic fistula (POPF), post pancreatectomy haemorrhage(PPH), delayed gastric emptying (DGE) ,sepsis. sometimes mortality.Hydrocortisone and indomethacin have been postulated to reduce post operative pancreatitis and thus PJ leak.
Methods: Between Jan 2018 - April 2019, 146 patients for Whipple's PD were included. Only high risk patient (n= 105)with >40% of acini (marker of soft pancreas) on frozen section of transection margin were randomized to intravenous (iv) hydrocortisone, per rectal (PR) indomethacin or placebo (3 groups ,35 in each group).
All patients received total 8 doses of iv treatment (8 hourly) and 6 doses of PR treatment
(12 hourly). 100 ml NS and glycerin suppository were the placebo drugs. Primary end-points were overall major complications(Clavien Dindo 2-5).
Results: Hydrocortisone group had less major complications compared to placebo
(overall 14.3% vs 40.0%; P value = 0.003).
POPF (8.6% vs 20%) and DGE (14.3% vs 22.9%) were also lower in hydrocortisone group
Indomethacin group did not reduce major complications compared to placebo (overall, 37.2% vs 40%; p value = 1.00). POPF (17.1% vs 20%) and PPH (11.4% vs 14.3%) Although, DGE is significantly less in indomethacin group (14.3% vs 22.9%; p value = 0.001).
30 day mortality was zero in all the groups.
Conclusions: Hydrocortisone treatment significantly reduces major postoperative complications in high risk patients after Whipple's PD whereas Indomethacin treatment does not as compared to placebo.
OP04-07 Preoperative Prediction of Clinically Relevant Postoperative Pancreatic Fistula after Pancreaticoduodenectomy
Hirohisa Kitagawa, Japan

H. Kitagawa1,2, K. Hashida1, J. Muto1, K. Hamai1, M. Okabe1, K. Kawamoto1, I. Makino2, F. Toshima3, D. Inoue3
1Surgery, Kurashiki Central Hospital, Japan, 2Surgery, Kanazawa University, Japan, 3Radiology, Kanazawa University, Japan

Introductions: Clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreaticoduodenectomy (PD) can complicate postoperative course and it is stressful for surgeons. It's ideal to predict CR-POPF preoperatively and take precautions against serious complications in advance.
The aim of this study was to clarify objective and predictive preoperative parameters of CR-POPF.
Methods: A consecutive cohort of PD patients from 2011 to 2017 were identified from a prospectively collected institutional database. CR-POPF was diagnosed according to the Revised 2016 ISGPS classification. Surgery-related factors (age, sex, bleeding volume and disease), morphologic and imaging parameters by CT (main pancreatic duct diameter, pancreatic parenchymal thickness, ratio of pancreatic parenchyma diameter to main pancreatic duct diameter above portal vein in CT axial images [P/D ratio], and contrast effect of pancrearic parenchyma [CE] in various phase) and signal intensity of MRI (in T1WI, T2WI, arterial phase, portal phase, and late phase) were examined.
Results: 104 patients were included in the analysis. Overall, 32 (31%) patients developed a CR-POPF. Multivariate analysis was performed on significant factors by univariate analysis, the P/D ratio (odds ratio [OR] 3.77, 95% confidence interval 1.27 to 11.93; P=0.017) and CE in late phase ([OR] 4.23, 95% confidence interval 1.20 to 18.11; P=0.024) were significant factors.
Conclusions: P/D ratio and CE in late phase were significant factors for predicting CR-POPF preoperatively. For high-risk patients, effective measures should be considered to prevent serious complications caused by POPF preoperatively.
OP04-08 Comparison of Outcomes of Minimally Invasive versus Open Pancreaticoduodenectomy after Neoadjuvant Therapy in a National Cohort of Patients with Pancreatic Adenocarcinoma
Amr Al Abbas, United States

A. Al Abbas, C.A. Hester, J. Yan, H. Zhu, H.J. Zeh III, P.M. Polanco
Surgery, University of Texas Southwestern Medical Center, United States

Introduction: The recently published “Miami International Evidence-based Guidelines on Minimally Invasive Pancreatic Resection”, endorsed by IHPBA, noted lack of evidence on which pancreaticoduodenectomy (PD) approach was optimal for pancreatic adenocarcinoma (PDAC) patients post-neoadjuvant therapy (NAT). We aim to compare 30-day-outcomes in PDAC patients that underwent open (OPD) versus minimally-invasive approach (MIPD) post-NAT.
Methods: PDAC patients that underwent NAT followed by either MIPD (laparoscopic/robotic) or OPD were identified in the U.S. Procedure-Targeted-Pancreatectomy NSQIP dataset (2014-2017). Preoperative and postoperative parameters were compared among both approaches. Subsequent analysis based on operative approach and NAT modality (chemotherapy alone or chemoradiation) was performed using multiple logistic regression models.
Results: Of 2428 patients with PDAC that received NAT, 2219 (91.4%) and 209 (8.6%) underwent OPD and MIPD respectively. MIPD patients were more likely to receive chemotherapy alone (70.8 vs 57.5%,P< 0.001), less likely to undergo vascular resection (21.2vs37.5%, P< 0.001), had longer OR-Time(P=0.043), and shorter Length-Of-Stay
(P< 0.001). For patients undergoing chemoradiation, MIPD was independently predictive of lower incidence of major complications (OR:0.22, P< 0.001). For patients who underwent chemotherapy alone, MIPD was independently predictive of a lower incidence of minor complications (OR:0.76, P=0.032) and a shorter LOS (Estimate:-2days,
P=< 0.001).There were no significant differences in mortality, Delayed-Gastric-Emptying (DGE), and Clinically-relevant-pancreatic-fistula(CR-POPF) among operative approaches and NAT modalities.
Conclusions: MIPD is safe and feasible in PDAC patients undergoing NAT in centers with expertise in this approach. No differences in mortality,DGE and CR-POF were found. MIPD was associated with a lower incidence of major complications in the chemoradiation group and shorter LOS with lower incidence of minor complications in the chemotherapy-only group.
[Multivariable Analysis for Major Complications by Operative Approach and Neoadjuvant Therapy]
OP04-11 Assessment of Glucose Metabolism Alterations after Partial Pancreatectomy Using Biochemical Markers: A Prospective Observational Study
Jun Suh Lee, Korea, Republic of

J.S. Lee, Y.-S. Yoon, B. Lee, J.Y. Cho, H.-S. Han
Division of Hepatobiliary and Pancreatic Surgery, Department of Surgery, Seoul National University Bundang Hospital, Korea, Republic of

Introduction: Most previous studies on glucose metabolism after pancreatectomy used medical records as diagnostic criteria of DM. This study aimed to evaluate the incidence and characteristics of new-onset DM (NODM) and worsened preexisting DM after pancreatectomy using serial assessment of biochemical markers in a prospective cohort.
Method: Data was prospectively collected for 224 patients who received PD (n=149) and DP (n=75) between 2015 and 2018. Diabetes related parameters were assessed preoperatively and postoperatively (at 3 months and 1 year): oral glucose tolerance test, HbA1c, fasting insulin, and stimulated insulin. Homeostasis model assessment (HOMA) was calculated for IR (insulin resistance) and B (beta cell function).
Results: The incidence of NODM (14% vs. 45%, P=0.001) and worsened DM (21% vs. 60%, P< 0.001) was significantly higher after DP than PD at postoperative 1 year. There was more DM resolution after PD. (41% vs. 9%, P=< 0.001) BMI and type of surgery (DP) were risk factors of NODM, while only type of surgery (DP) was a risk factor of worsened DM. In DP patients without preoperative DM, those who developed NODM had a significantly lower preoperative HOMA-B level compared to those who did not. (P=0.035) PD patients who developed NODM showed a sustained decrease in HOMA-B postoperatively, whereas those who did not showed a plateau, after an initial decrease.
Conclusions: DP had higher risk of NODM development and DM worsening that PD. Patient education and surveillance for the development of DM after pancreatectomy should be tailored according to type of resection.
OP04-12 Impact of Radical Pancreaticoduodenectomy with Portal Vein Resection and Extensive Surrounding Soft Tissues on the Long-term Outcomes for Pancreatic Head Cancer
Akio Saiura, Japan

A. Saiura1,2, H. Ito2, A. Oba2, Y. Inoue2, Y. Mise1, Y. Takahashi2, Y. Ono2, T. Sato2, R. Yoshioka1
1Hepato-Biliary-Pancreatic Surgery, Juntendo University Graduate School of Medicine, Japan, 2Hepato-Biliary-Pancreatic Surgery, Cancer Institute Hospital, Japan

Background: Pancreaticoduodenectomy (PD) with portal vein PV resection (PVR) is a standard operation for pancreatic ductal adenocarcinoma (PDAC) with PV invasion; however, the positive margin rates remain high. We hypothesized that radical pancreaticoduodenectomy (RPD) in which soft tissue around the PV is resected enbloc could enhance oncological clearance. Herein, we describe our RPD and address the short- and long-term outcomes compared to standard PD with PVR.
Method: The study included 268 consecutive patients who underwent PD with PVR using anterior artery-first approach. While, the PV was skeletonized with the surrounding soft tissue dissected in the standard PD with PVR (n = 177), the retro-pancreatic segment of PV was resected enbloc with its surrounding soft tissue during the RPD (n = 91). The extent of lymphadenectomy was not different between the procedures.
Results: R0 resection was achieved in 80% of patients in the RPD group, compared with 66% in the PD group (P = 0.011), while the perioperative outcomes were comparable between groups. The median recurrence-free survival (RFS) and overall survival (OS) were 17 months and 31 months, respectively, for the RPD group, compared to 11 months and 21 months for the PD group, (P = 0.004 for RFS and P = 0.003 for OS).
Conclusion: We described a novel, radical operation for locally advanced PDAC. Our RPD is safe and feasible, and it enhances local disease control resulting in improved OS. Further prospective evaluation of RPD is warranted in the setting of current multidisciplinary management.
OP04-13 Neoadjuvant Treatment Mitigates the Survival Impact of Major Complications after Resection of Pancreatic Adenocarcinoma
Timothy E. Newhook, United States

T. Newhook, L. Prakash, M. Bruno, W. Dewhurst, N. Ikoma, M. Kim, J.E. Lee, M. Katz, C.W. Tzeng
MD Anderson Cancer Center, United States

Postoperative major complications (PMCs) may prevent multimodality therapy (MMT) for pancreatic ductal adenocarcinoma (PDAC) patients by delaying adjuvant therapy (AT) following surgery-first (SF) sequencing. We hypothesized that neoadjuvant therapy (NT) mitigates the detrimental effect of PMCs on outcomes of resected patients.
Characteristics of consecutive resected PDAC patients 7/2011-10/2018 were abstracted from a prospective database. PMCs were defined at 90-days as ACCORDION Grade ≥3. Overall survival (OS) was compared between patients with and without PMCs.
Of 373 patients, most underwent NT (75%). PMCs occurred in 22% of SF and 20% of NT patients (p=0.71). Most went on to receive some form of AT (90% SF vs. 70% NT,p< 0.001). Median OS for NT and SF patients was 46 vs. 36 months (p=0.037). PMCs negatively impacted OS, with median OS 59 months for NT(-)PMC, 34 months for NT(+)PMC, 45 months for SF(-)PMC, and 20 months for SF(+)PMC (p< 0.001;Fig. 1A). There was a trend toward worse OS in NT(+)PMCs (p=0.06, Figure 1B). PMCs were not independent predictors of OS for NT patients. However, after adjustment for clinical classification, treatment sequencing, tumor size, and margin status, PMCs were independently-associated with OS (HR-1.60,p=0.010) among all patients, along with perineural invasion (HR-1.83, p=0.024), nodal positivity (HR-2.1,p< 0.001), and AT (HR-0.69,p=0.039).
The deleterious effects of PMCs on OS for PDAC patients may be mitigated by NT. NT sequencing should be routinely considered given the significant risk of post-pancreatectomy morbidity.
[Figure 1]
OP04-15 Clinical Relevance Between Survival Outcomes and Invasion of Splenic Vessels in Pancreatic Body or Tail Adenocarcinoma
Jae Seung Kang, Korea, Republic of

J.S. Kang, Y.J. Choi, Y. Byun, Y. Han, E. Kim, H. Kim, W. Kwon, J.-Y. Jang
Surgery and Cancer Research Institute, Seoul National University Hospital, Seoul National University College of Medicine, Korea, Republic of

Introduction: Little was known about clinical impact of splenic vessels invasion (SpVI) of pancreatic body or tail adenocarcinoma in terms of survival outcomes. This study was to compare the survival outcomes between pancreatic adenocarcinomas (PDACs) with SpVI and those with no invasion, and to investigate the prognostic factors associated with adverse outcomes.
Methods: Between 2005 and 2018, patients who underwent distal pancreatectomy were enrolled. Patients who underwent neoadjuvant chemotherapy were excluded. Degree of SpVI was categorized with three groups (Group 1, no invasion; Group 2, 0 - 180 degree; Group 3, ≥180 degree) and formation of collateral vessels was investigated in preoperative computed tomography. Clinical variables, postoperative surgical outcomes, and survival outcomes were evaluated. Multivariate Cox-proportional analysis was performed for evaluating the prognostic factors.
Results: Total 249 patients were included. Operation time was longer (185 vs. 159 min, P=0.001) and intraoperative blood loss (415 vs. 278 mL, P=0.003) was higher in SpVI patients. Tumor size was larger (3.9 vs. 2.9cm, P=0.001) in SpVI patients, but the number of metastatic lymph nodes were comparable (1.7 vs. 1.4, P=0.241). 5-year overall survival rate was significantly different among three groups (Group 1, 38.4%; Group 2, 16.8%; Group 3, 9.7%, P< 0.001). In the Cox-proportional analysis, adjuvant treatment, R0 resection, SpVI, and collateral vessels formation were independent prognostic factors in survival outcome.
Conclusions: SpVI was associated with adverse survival outcomes in PDAC. Different approach such as neoadjuvant treatment would be needed in patients with SpVI invasion.
OP04-16 Changes in the Perioperative and Postoperative Long-term Quality of Life after Total Pancreatectomy
Moon Young Oh, Korea, Republic of

M.Y. Oh, E.J. Kim, W. Kwon, H. Kim, Y.H. Byun, Y.J. Choi, J.S. Kang, Y. Han, J.-Y. Jang
Department of Surgery, Seoul National University Hospital, Korea, Republic of

Background: Quality of Life (QoL) is widely known to be poor after total pancreatectomy. This study was designed to evaluate the short-term and long-term consequences of endocrine and exocrine insufficiency and their associated effects on QoL and nutritional status.
Methods: Prospective data was collected from patients who underwent total pancreatectomy at Seoul National University Hospital during an interval of 4 years and followed up for at least 1 year. QoL, and nutritional status were assessed by administering validated questionnaires (EORTC QLQ C-30, PAN26, GIQLI, MNA), preoperatively and 3, 12 months postoperatively.
Results: A total of 30 patients were eligible for the study. 3 months after receiving total pancreatectomy, the global heath score (GHS) showed no significant difference (preoperatively 57.2 vs. 3 months postoperatively 68.3; P=0.119). By the 1st postoperative year, the GHS still showed no significant difference (preoperatively 57 vs. 1 year postoperatively 52.4; P=0.2) and no significant differences in most of the QoL categories. However, poor physical function (79.2 vs. 67.6; P=0.01), digestive difficulties (14.9 vs. 36.9; P=0.03) and altered bowel habits (9.2 vs. 25.6; P=0.03) continued even 1 year after surgery.
Conclusion: The overall QoL score after total pancreatectomy was comparable to the preoperative QoL score. Some symptoms after total pancreatectomy significantly worsen after 3 months postoperatively, but then improve to a comparable level 1 year after surgery. Because some symptoms persist even after time has passed, supportive management is needed for total pancreatectomy patients, including nutritional support with pancreatic enzyme replacement and education for diabetes and diet.
OP04-17 Comparative Long-term Outcomes for Pancreatic Volume Change, Nutritional Status, and Incidence of New-onset Diabetes between Pancreatogastrostomy and Pancreatojejunostomy after Pancreaticoduodenectomy
Bong Jun Kwak, Korea, Republic of

B.J. Kwak1, H.J. Choi2, Y.K. You2,2, T.H. Hong2
1Division of Hepatobiliary, Pancreas, and Abdominal Organ Transplant, Department of Surgery, Incheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Korea, Republic of, 2Department of Hepatobiliary and Pancreas Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Korea, Republic of

Introduction: The difference in volume change in a pancreatic remnant according to the type of pancreaticoenterostomy after pancreaticoduodenectomy (PD) for long-term follow-up is unknown. This study aimed to compare serial pancreatic volume changes in pancreatic remnants between pancreatogastrostomy (PG) and pancreatojejunostomy (PJ) after PD and to evaluate the difference in general nutritional status and incidence of NODM between PG and PJ.
Methods: This study enrolled 115 patients who had survived for more than three years after PD. They were divided into the PG group and the PJ group. Their clinicopathologic factors were collected and analyzed. We calculated serial pancreas volume and pancreatic duct size precisely from preoperative stage to five years after surgery by image-processing software. Consecutive changes of albumin and BMI as related to general nutritional status were compared. Postoperative NODM was evaluated.
Results: Most patient demographics were not significantly different between the PG group (n=45) and PJ group (n=70). There was no significant difference in volume reduction between the groups from postoperative one month to five years (PG group −18.21±14.66 mL versus PJ group −14.43±13.05 mL, P=0.209). There was no significant difference in the change of total serum albumin and BMI between the groups for five years after surgery. The incidence of NODM was not significantly different between the groups (P=0.995).
Conclusions: PG and PJ following PD induced similar pancreatic volume reduction during long-term follow-up. There was no difference in general nutritional status or incidence of NODM between the groups after PD.
OP04-18 Comparisons of Short-term and Long-term Outcomes Between Open and Laparoscopic Distal Pancreatectomy in Patients with Pancreatic Ductal Adenocarcinoma
Jung Min Lee, Korea, Republic of

J.M. Lee, J.S. Kang, Y. Byun, Y.J. Choi, Y. Han, H. Kim, W. Kwon, J.-Y. Jang
Department of Surgery and Cancer Research Institute, Seoul National University College of Medicine, Korea, Republic of

Safety and feasibility of laparoscopic distal pancreatectomy (LDP) in regards to the pancreatic adenocarcinoma (PDAC) were not well-known. The present study aimed to compare the short-term and long-term outcomes of LDP with those of open distal pancreatectomy (ODP).
This was a retrospective study with prospectively collected medical data. Between 2009 and 2017, patients who underwent distal pancreatectomy and pathologically confirmed as PDAC were enrolled. Clinical and pathologic variables were investigated. To reduce selection bias, 1:1 propensity score matching (PSM) was performed with T and N stage of 8th American Joint Committee on Cancer staging system. Survival outcomes and cumulative recurrence rates were calculated with Kaplan-Meier method.
Total 210 patients were enrolled. LDPs and ODPs were performed in 35 patients (16.7%) and 175 patients (83.3%), respectively. After 1:1 PSM, age, sex, underlying diseases were comparable between two groups. In terms of short-term outcomes, operation time (128 vs. 164 minute, P=0.001) and postoperative hospital stay (11.1 vs. 16.5 days, P=0.011) were significantly different between two groups. Tumor size (3.2 vs. 3.1 cm, P=0.889), number of harvested lymph nodes (12.6 vs. 14.4, P=0.365), and R0 resection rates (91.4 vs. 80.0%, P=0.172) were comparable. 5-year overall survival rates (26.4 vs. 24.6%, P=0.742) and cumulative recurrence rates (56.3 vs. 61.4%, P=0.582) were comparable between two groups.
LDP has similar or better perioperative outcomes (operation time, postoperative hospital stay) and shows similar survival outcomes, and recurrence patterns in PDAC patients, compared with ODP. LDP is a safe and feasible procedure in PDAC patients.
OP04-19 Somatostatin Prevents Clinically Relevant Pancreatic Fistula in Intermediate Risk Patients after Pancreaticoduodenectomy (SPEED): A Multi-center, Randomized, Controlled Study
Zhe Cao, China

Z. Cao1, J. Qiu1, T. Zhang1, B. Sun2, R. Qin3, R. Chen4, Y. Miao5, W. Lou6, Y. Zhao1
1General Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, China, 2Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Harbin Medical University, China, 3Biliary-Pancreatic Surgery, Affiliated Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, China, 4Pancreatobiliary Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, China, 5General Surgery, The First Affiliated Hospital, Nanjing Medical University, China, 6Pancreatic Surgery, Zhong Shan Hospital, Fudan University, China

Introduction: Post-operative pancreatic fistula (POPF) remains the lethal complication after pancreaticoduodenectomy, and the objective of the study is evaluating the preventive effect of somatostatin on POPF in intermediate risk patients.
Methods: A multi-center, randomized, controlled study was conducted in six high-volume pancreas centers in China between June 2018 and April 2019. Patients undergoing pancreaticoduodenectomy with intermediate risk of POPF were enrolled. Patients were randomly assigned to somatostatin group (intravenous somatostatin of 250µg/h for 120 hours) and control group. The primary endpoint was clinically relevant POPF (CR-POPF) (according to 2016 International Study Group on Pancreatic Fistula criteria). This trial was registered with Clinical Trial (NCT03349424).
Results: 205 patients were enrolled and 99 in somatostatin group and 100 in control group were included for final analysis. The rate of CR-POPF in somatostatin group decreased significantly (13% vs 25%, p=0.032), both in open and laparoscopic pancreaticoduodenectomy. But the rates of overall POPF (65% vs 69%, p=0.51) and biochemical leak (52% vs 44%, p=0.29) were not significantly different. Medical costs (¥115069 vs ¥115803, p=0.92) and other complications: biliary fistula (6% vs 6%, p=0.99), abdominal infection (19% vs 18%, p=0.83), chylous fistula (5% vs 4%, p=0.75), late postoperative hemorrhage (7% vs 12%, p=0.24) had no significant difference. However, the somatostatin group had higher rate of delayed gastric emptying (33% vs 21%, p=0.0504).
Conclusion: In patients with intermediate risk of POPF after pancreaticoduodenectomy, prophylactic use of somatostatin can reduce the CR-POPF, but seems to increase the rate of delayed gastric emptying.
 Overall (n=199)somatostatin group (n=99)control group (n=100)
Age, mean (SD), yrs58.857.92 (11.4)59.18 (10.7)
Male sex123(62%)57 (57%)66 (66%)
BMI, mean (SD) (kg/m2)22.922.58 (3.2)23.26 (3.2)
Hypertension inmedical history56(28%)56(28%)32 (32%)
Diabetes mellitus in medical history33(17%)16 (16%)17 (17%)
Chronic pancreatitis in medical history2(1%)2 (2%)0(0%)
Acute pancreatitis in medical history5(3%)0(0%)5 (5%)
[The baseline characteristics of study participants.]


[The rates of POPF and other complications in somatostatin and control group.]
OP04-20 The European Registry for Minimally Invasive Pancreatic Surgery (E-MIPS): First Year Experience
Nicky van der Heijde, Netherlands

N. van der Heijde1,2, F. Vissers1,2, F. Can3, T. Hackert4, I. Khatkov5, O. Saint-Marc6, G. Zimmitti2, M. Besselink1, M. Abu Hilal2,7, European consortium of Minimally Invasive Pancreatic Surgery (E-MIPS)
1Department of Surgery, Amsterdam University Medical Center, Location AMC, Netherlands, 2Department of Surgery, Fondazione Poliambulanza Hospital, Italy, 3Department of Surgery, Lokman Hekim University School of Medicine, Turkey, 4Department of General, Visceral, and Transplantation Surgery, Heidelberg University Hospital, Germany, 5Department of Surgery, Moscow Clinical Scientific Center, Russian Federation, 6Department of Surgery, Centre Hospitalier Régional Orleans, France, 7Department of Surgery, University Hospital Southampton NHS, United Kingdom

Introduction: The European-African Hepato-Pancreato-Biliary Association (E-AHPBA) has endorsed the European consortium on Minimally Invasive Pancreatic Surgery (E-MIPS) to set up a registry which aims to collect data of minimally invasive pancreatic surgery (MIPS) in all low- and high-volume centers across Europe. The aim is to monitor and report on safety and quality outcomes of MIPS in daily clinical practice.
Methods: This is a pan-European, multicenter prospective observational cohort study, including data of the first year (2019) of the E-MIPS registry. All patients undergoing MIPS in the participating centers are included. Main study parameters are patient demographics, perioperative- and oncological outcomes.
Results: A total of 398 patients from 38 centers in 15 countries were included, with a median (inter quartile range) volume of 11 (8-20) for MI-distal pancreatectomy. MI-pancreatoduodenectomy was performed in 23 centers, with a median (IQR) of 8 (2-20). There were 31 (81.6%) low volume (< 20 MIPD annually) centers and 7 (18.4%) high-volume centers. Laparoscopy was the most frequent approach (n=245, 61.3%), followed by robotic (n=134, 33.5%) and hybrid-laparoscopic (n=19, 4.8%). Overall, 240 patients (60%) were operated for a malignancy, of which 13 (5.4%) received a type of neoadjuvant treatment. The 90-day mortality rate was 2.6% (n=10). Table 1 shows summarized outcomes divided between the five procedures performed most often.
Conclusion: This is the first overview of collected data from all centers in the E-MIPS registry. Due to the large scale, this registry provides insight into the current MIPS practice in Europe.
 Robotic pancreatoduodenectomy (n=70)laparoscopic pancreatoduodenectomy (n=61)Hybrid-laparoscopic pancreatoduodenectomy (n=18)Robotic distal pancreatectomy (n=70)laparoscopic distal pancreatectomy (n=144)
Age, years, mean (SD)64 (11)63 (12)68 (10)60 (16)63 (15)
Conversion, n (%)5 (7.1)11 (13.4)6 (33.3)5 (7.1)20 (14.1)
Length of stay, med (IQR)10 (8-16)11 (8-16)8 (6-12)7 (6-10)6 (5-10)
POPF grade B/C, n (%)11 (15.7)13 (15.9)2 (11.1)9 (15.5)32 (22.7)
PPH grade B/C, n (%)5 (7.1)6 (7.3)0 (0)3 (5.2)5 (3.5)
Reoperation ≤30 days, n (%)8 (11.4)8 (9.8)1 (5.6)1 (1.6)7 (4.9)
Readmission ≤30 days, n (%)5 (7.1)10 (12.2)4 (22.2)7 (11.5)23 (16.0)
R0 resection, n (%)*39 (68.4)59 (83.1)10 (71.4)29 (90.6)65 (82.3)
90-day mortality, n (%)2 (2.9)7 (8.5)0 (0)1 (1.6)1 (0.7)
[Baseline characteristics and perioperative outcomes of the European minimally invasive pancreatic surgery (E-MIPS)registry]
OP04-21 Increased Operative Difficulty and Poorer Outcomes after Multiple Endotherapies for Chronic Pancreatitis: An Analysis of 48 Consecutive Frey's Procedures
Prasad Pande, India

P. Pande, G. Desai, R. Narkhede, P. Wagle
Lilavati Hospital and Research Centre, India

Introduction: Chronic pancreatitis (CP) is being increasingly treated by endotherapy as part of the step-up approach. Multiple sessions of endotherapy have led to shrinking indications and delayed referrals for surgery. This study analyses the impact of multiple endotherapy sessions on difficulty of Frey's procedure and outcomes of surgery.
Material and Methods: This prospective study included 48 consecutive Frey's procedures done for CP between 2016 and 2019 at our tertiary hepatopancreatobiliary centre. Demographic data, duration of CP, number of endotherapies [≥3 (group A) or < 3 (group B)], operative difficulty [operative time, intra-operative blood loss], pain relief [Visual Analog Scale (VAS)], and quality of life [EORTC-QLQ-C30 questionnaire (QoL score)] at 6 months were recorded.
Results: 28 (58.33%) out of 48 patients were in group A, and 20 (41.67%) in group B. The mean operative time was 153.22±24.1 minutes in group A vs 138.33±15.29 minutes in group B (p< 0.05). The mean blood loss was 156.11±43.28 mL in group A compared to 116.04±34.42 mL in group B (p< 0.01). Improvement in VAS at 6 months was 3.22±0.83 in group A and 4.79±1.58 in group B (p< 0.01). Improvement in QoL score at 6 months was 36.44±9.34 in group A and 35.62±12.42 in group B (p=0.85).
Conclusion: Multiple endotherapy sessions increase the operative time and intra-operative blood loss, thus increasing the operative difficulty, in addition to giving poorer pain relief. Early referral for surgery without subjecting patient to multiple endotherapies would improve operative and post-operative outcomes.
Parameter≥3endotherapies (group A)<3 endotherapies (group B)p-value
Number of patients28 (58.33%)20 (41.67%) 
Mean duration of surgery (minutes)153.22 ± 24.1138.33 ± 15.29<0.05*
Mean blood loss (mL)156.11 ± 43.28116.04 ± 34.42<0.01*
Change in VAS at 6 months3.22 ± 0.834.79 ± 1.58<0.01*
Change in QoL score at 6 months36.44 ± 9.3435.62 ± 12.420.85
[Correlation of number of endotherapies with operative difficulty and outcomes]
OP04-22 Predictive Value of Elevated CA 19-9 for Positive Resection Margins after Pancreaticoduodenectomy for Pancreatic Tumors
Namita Chavan, India

N. Chavan, P. Pande, G. Desai, R. Shah, P. Jagannath
Lilavati Hospital and Research Centre, India

Introduction: Elevated Carbohydrate antigen (CA) 19-9 levels have predicted poor prognosis and decreased survivals after pancreaticoduodenectomy for pancreatic head carcinoma. This study analyzes the predictive value of elevated CA 19-9 for positive resection margin in these cases.
Material and methods: Retrospective analysis of prospectively entered data from 2011 to 2019 revealed 202 cases at our specialized hepatopancreatobiliary centre. Demographic details, imaging findings, liver function tests and pre-operative tumor markers were recorded. Histopathology data regarding tumor and nodal status and resection margins were recorded as per Leeds protocol. Positive margins were considered as < 1mm from the tumor.
Results: 106 of 202 (52.4%) patients had CA 19-9 >100 U/mL (Normal: < 37 U/mL) in absence of jaundice. 63 (31.2%) patients had node positive disease, and 78 (38.6%) had lymphovascular/perineural invasion. Retroperitoneal (SMA) margin was positive in 16 (15.1%) patients with elevated CA 19-9 and 4 (4.1%) patients with normal CA 19-9. SMV margin was positive in 38 (35.8%) patients with elevated CA 19-9 and 11 (11.4%) patients with normal CA 19-9. Pancreatic ductal margin was positive in 3 (3.1%) patients with elevated Ca 19-9. 57 (53.7%) patients with elevated CA 19-9 and 15 (15.6%) patients with normal Ca 19-9 had positive margins. This difference is statistically significant (p< 0.05) using student's t-test.
Conclusion: Elevated CA 19-9 level is a strong predictor of margin positivity after pancreaticoduodenectomy irrespective of T stage. These cases may be better suited for neoadjuvant therapy and needs evaluation in a prospective study.
ParameterCA 19-9 >100 U/mLCA 19-9 <100 U/mLp-value
Number of patients106 (52.4%)96 (47.5%) 
N+ disease63 (31.2%)21 (10.3%)<0.05*
Median LN ratio0.33±0.260.14±0.12<0.05*
LVI/PNI78 (38.6%)35 (17.3%)<0.05*
Positive SMA margin16 (15.1%)4 (4.1%)<0.05*
Positive SMV margin38 (35.8%)11 (11.4%)<0.05*
Positive duct margin3 (3.1%)0 
Positive margin status57 (53.7%)15 (15.6%)<0.05*
[Histopathology and margin status in pancreaticoduodenectomy]
OP04-23 Relevance of Celiac Axis Stenosis in Pancreatoduodenectomy
Mohammed Al-Saeedi, Germany

M. Al-Saeedi1, H. Sauer1, J. Koch1, L. Frank-Moldzio1, P. Mayer2, T. Hackert1, T. Bruckner3, M.W. Büchler1, O. Strobel1
1Department of General, Visceral and Transplantation Surgery, University Hospital of Heidelberg, Germany, 2Department of Diagnostic and Interventional Radiology, University Hospital of Heidelberg, Germany, 3Institute of Medical Biometry and Informatics, University Hospital of Heidelberg, Germany

Introduction: Celiac axis stenosis (CAS) may result in enhanced risk of ischemic complications during pancreatoduodenectomy. However, the prevalence and relevance of CAS remains unknown.
Methods: All patients undergoing partial or total pancreatoduodenectomy from 2014 to 2017 after preoperative computed tomography (CT) with arterial phase were identified from a prospective database. Preoperative CT scans were evaluated for CAS. Postoperative complications were assessed.
Results: Of 998 patients 273 (27.4%) had CAS. The degree of radiological CAS was 30-50% in 7.7%, 50-80% in 8.3%, and 80-100% in 1.3% of patients. CAS (of any degree above 30%) was associated with increased morbidity including intra-abdominal collections (p=0.022), gastric ischemia (p=0.001), liver ischemia (p< 0.001), pancreatic fistula (p=0.003), sepsis (p=0.007), as well as with longer hospital stay (p< 0.001) and ICU stay (p=0.016). Patients with CAS required significantly more surgical reinterventions (p=0.001) including gastric (p< 0.001) and pancreas reoperations (p=0.007). Rate and severity of complications including liver ischemia and pancreatic fistula increased with higher degree of CAS. Among patients with 80-100% stenosis, both grade B/C pancreatic fistula and at least moderate liver failure occurred in 46% of patients. Multivariable analyses confirmed CAS as independent risk factor both for liver ischemia (p=0.010) and for pancreatic fistula (p=0.007).
Conclusion: CAS is common and represents an underestimated risk for relevant complications after pancreatoduodenectomy. Already a radiological stenosis of 30-50% is associated with increased risk of morbidity. Precise radiological assessment of the celiac axis may help to identify risk, to address relevant CAS, and, thus, to avoid postoperative complications.
OP04-24 Short-Term Clinical Outcomes after Total Pancreatectomy: A Prospective Multicenter European Snapshot Study
Anouk Latenstein, Netherlands

A. Latenstein1, L. Scholten1, M. Erkan2, J. Kleeff3, M. Lesurtel4, M. de Pastena5, A. Halimi6, M. Besselink1, J. Ramia-Angel1, Scientific and Research Committee of the E-AHPBA
1Amsterdam UMC, Location AMC, Netherlands, 2Koç University Hospital, Turkey, 3Martin Luther University, Germany, 4University Hospital of Zurich, Switzerland, 5University and Hospital Trust of Verona, Italy, 6Karolinska University Hospital, Sweden

Introduction: Prospective multicenter studies on clinical outcomes after total pancreatectomy (TP) are not reflecting current practice due to long study periods and inclusion of mainly high-volume centers. The aim of this prospective multicenter European snapshot study is to assess short-term clinical outcomes after elective TP.
Methods: Patients who underwent elective TP for malignant or benign disease between June 2018 and June 2019 were prospectively included from 42 hospitals. Hospitals were divided based on annual volume (low 1-9, medium 10-19, and high ≥20 TPs). Variables associated with major postoperative complications (Clavien Dindo ≥3) and 90-day mortality were assessed in multivariable logistic regression.
Results: In total, 276 patients underwent TP, mostly for malignant disease (73%). Minimally invasive TP was performed in 11 (4%) patients. Major postoperative complications occurred in 25% and predictors were ASA score ≥3 (OR 2.41 [95%CI 1.27-4.55] p=0.007), blood loss (OR 1.00 [95%CI 1.00-1.00] p=0.009), and low-volume centers (OR 2.28 [95%CI 1.15-4.52], p=0.019). Median hospital stay was 12 days (IQR 9-18) and the 90-day readmission rate was 14%. The 30-day and 90-day mortality rates were 4% and 8%, respectively. In multivariable analysis, only age (OR 1.07 [95%CI 1.02-1.13], p=0.008), BMI (OR 1.11 [95%CI 1.01-1.23], p=0.039) and hospitals with 1-9 patients (reference ≥20 patients, OR 4.78 [95%CI 1.56-14.71], p=0.006) were predictors for 30-day mortality.
Conclusion: This prospective multicenter study found 25% major postoperative complications and 8% 90-day mortality after TP and evidence to suggest that also for TP a higher volume may reduce postoperative mortality.
OP04-25 Minimally Invasive Pancreaticoduodenectomy (MIPD) at a Tertiary Centre over 2 Decades- "Lessons Learnt & Techniques Modified"
S Srivatsan Gurumurthy, India

S. Srivatsan Gurumurthy, M. Srinivasan, N. Anand Vijai, P. Senthilnathan, C. Palanivelu
Dept. of HPB, Minimally Invasive Surgery & Liver Transplant, Gem Hospital, India

Introduction: In 1998, the first totally Laparoscopic pancreaticoduodenectomy (LPD) for periampullary carcinoma was performed in our institute. 22 years since then ,MIPD at our institute has undergone several technical modifications, the recent addition ,over the last 2 years being Robotic Pancreaticoduodenectomy (RPD). Lessons learnt and techniques modified in this journey over the past 2 decades have been presented.
Methods: A retrospective analysis of prospectively maintained database was carried out for all MIPD cases (418 LPD & 42 RPD ) done at our institute from Dec 1998 to Dec 2019. A subgroup analyses was done to compare the outcomes of LPD in the first decade versus the second decade.
Results: The authors observed a reduction in operative time and estimated blood loss between the first decade and the second decade. The mean number of lymph nodes removed was higher in the second decade , though not statistically significant (p= 0.07) Estimated blood loss was comparable with laparoscopic and robotic PD (203.23+/- 84.37 vs 206.17 +/- 82.82, p=0.87) However robotic PD took a significantly longer time compared to LPD (392 +/- 124.26 vs 312 +/- 39.62 , p=0.02) in this series.
Conclusion: MIPD provides several advantages like decreased length of hospitalization, reduced blood loss and need for transfusion, more meticulous oncologic dissection and higher lymph node yield. Accumulation of surgical experience, better optics, improved energy sources and adoption of robotic technology have refined the outcomes of MIPD at our institute and worldwide, in recent times.
OP04-26 Risk Factors of Serious Postoperative Complications after Laparoscopic Pancreatoduodenectomy: Single Center Retrospective Study
Roman Izrailov, Russian Federation

I. Khatkov, R. Izrailov, O. Vasnev, M. Baychorov, P. Tytyunnik, A. Andrianov, M. Mikhnevich, P. Agami
High-Tech Surgery, Moscow Clinical Scientific Center, Russian Federation

Background: Two hundred and ninety laparoscopic pancreatoduodenectomies (LPDE) were performed by single surgical team.
Objective: to assess the short-term and long-term outcomes of LPDE and to reveal the risk factors for having the Clavien-Dindo IIIa-V complication.
Methods: 290 patients underwent LPDE during last 10 years. 169 were females and 131 were males. Mean age was 60 years (range 29-82). 244 patients were operated on because of malignancies and 46 because of benign diseases. Postoperative complications were graded according Clavien-Dindo classification. 109 perioperative factors and parameters including laboratory test results were analyzed in order to reveal their influence on developing of postoperative complications.
Results: Mean operative time was 419min and mean blood loss was 350cc. Total Clavien-Dindo IIIa-V complications (CRPOC) rate was 36.8% (107 patients). Among them IIIa - 16,2%, IIIb - 10.7%, IV - 3,8%, V - 6,2%. CRPOPF were diagnosed in total of 20,3% patients (15,5% grade B POPF, 4,8% grade C). DGE was diagnosed in 4,8% of patients. PPH complicates the postoperative course of 20 patients (6,9%). Sixteen of them had concomitant PF. Univariate analysis revealed that the age ≥65, male sex, the soft pancreatic tissue, absent of pancreatic hypertension, high intraoperative blood loss (≥400 cc), high BMI (≥30), diagnosis other than pancreatic adenocarcinoma, were independent risk factors for having serious complications.
Conclusion: age ≥65, male sex, the soft pancreatic tissue, absent of pancreatic hypertension, high intraoperative blood loss (≥400 cc), high BMI (≥30), diagnosis other than pancreatic adenocarcinoma were independent risk factors for having serious complications.
OP04-27 Neoadjuvant Chemoradiotherapy with S1 for Resectable Pancreatic Cancer
Masaru Matsumura, Japan

M. Matsumura1, S. Nemoto1, K. Tani1, Y. Ome2, G. Honda2, Y. Seyama1
1Hepato-Biliary-Pancreatic Surgery, Tokyo Metropolitan Cancer and Infectious Disease Center Komagome Hospital, Japan, 2Gastrointestinal Surgery, New Tokyo Hospital, Japan

Introduction: The efficacy of neoadjuvant chemoradiotherapy (NACRT) with S1 for resectable pancreatic invasive ductal adenocarcinoma (R-PDAC) has not been clarified.
Method: Clinical data of 124 patients who underwent NACRT with S1 for R-PDAC from October 2009 to December 2018 were reviewed and compared with those of 45 patients who underwent up-front surgery for R-PDAC on the same period. The regimen of NACRT was concomitant daily 80 - 120 mg of S1 with 1.8 Gy of radiation for 28 days up to 50.4 Gy.
Results: Levels of CA19-9 before treatment were not different between two groups (NACRT median 82.4 U/L [IQR 18.3 - 199.4] vs. up-front surgery: 72.6 U/L [27.5 - 167.8]). Disease control rate of NACRT was 88.7%. Resection rate was 84.7% in NACRT and 93.3% in up-front surgery(p=0.25). Morbidity of Clavien-Dindo classification ≥ Grade 3a was 11.4% and 21.4% (p=0.13). R0 resection rate was 98.1% vs. 83.3% (p< 0.01) and percentage of patients with pathological lymph node metastases was 24.5% vs. 54.8% (p< 0.01). Adjuvant chemotherapy was administered in 81.9% of NACRT and 81.0% of up-front surgery (p=0.89). Overall survival (OS) of intention to treat population were significantly different between two groups (median survival time [MST]: 49.0 months vs. 43.4 months, 3 year-OS 59.0% vs. 55% [p=0.39]). For cohorts with resection followed by adjuvant chemotherapy, NACRT has significantly better survival than up-front surgery (MST 83.7 months vs. 43.9 months, 3 year-OS 77.0% vs. 61.0% [p=0.03]).
Conclusions: NACRT with S1 is feasible strategy for R-PDAC.
[OS of Intention to treat population and cohorts with resection and adjuvant chemotherapy]
OP04-28 Impact of Neoadjuvant Therapy on Postoperative Pancreatic Fistula: A Systematic Review and Meta-analysis
June Oo, Australia

J. Oo1, S. Kamarajah2,3, J. Bundred4, C. Boyle2, S. Pandanaboyana2, B. Loveday1,5,6
1Peter MacCallum Cancer Centre, Australia, 2Freeman Hospital, United Kingdom, 3Newcastle University Trust Hospitals, United Kingdom, 4University of Birmingham, United Kingdom, 5Royal Melbourne Hospital, Australia, 6University of Auckland, New Zealand

Introduction: The use of neoadjuvant therapy (NAT) for pancreatic cancer is increasing, although its impact on postoperative pancreatic fistula (POPF) is variably reported. This systematic review and meta-analysis aimed to assess the impact of NAT on POPF.
Methods: A systematic literature search until October 2019 identified studies reporting POPF following NAT (radiotherapy, chemotherapy or chemoradiotherapy) vs. upfront resection. The primary outcome was overall POPF. Secondary outcomes included Grade B/C POPF, delayed gastric emptying (DGE), postoperative pancreatic haemorrhage (PPH), and overall and major complications.
Results: The search identified 24 studies: pancreaticoduodenectomy (PD), 19 studies (n=19,893); distal pancreatectomy (DP), 5 studies (n=477). Local staging was reported in 17 studies, with borderline resectable and locally advanced disease comprising 6% (0 - 100%) and 1% (0 - 33%) of the population, respectively. For PD, any NAT was significantly associated with lower rates of overall POPF (OR: 0.57, p< 0.001) and Grade B/C POPF (OR: 0.55, p< 0.001). In DP, NAT was not associated with significantly lower rates of overall or Grade B/C POPF.
Conclusion: NAT is associated with significantly lower rates of POPF after PD but not after DP. Further studies are required to determine whether NAT should be added to POPF risk calculators.
OP04-29 Comparison with Short- and Long-term Neoadjuvant Chemoradiotherapy for Resectable and Borderline Resectable Pancreatic Adenocarcinoma
Hironobu Suto, Japan

H. Suto, K. Okano, M. Oshima, Y. Ando, H. Matsukawa, Y. Suzuki
Gastroenterological Surgery, Kagawa University, Japan

Introduction: The indications of preoperative treatment for resectable (R) borderline resectable (BR) pancreatic ductal adenocarcinoma (PDAC) are still obscure, and the protocol has not yet been standardized.
Method: The patients were divided into R, BR with venous involvement (BR-V) according to the 2019 NCCN guidelines. Between September 2009 and May 2016, short neoadjuvant chemoradiotherapy (NACRT) (3Gy x 10fr.+S-1) in 2 weeks was given to patients with R(n=33), BR-V(n=19). Subsequently, since June 2016, long NACRT (2Gy x 25fr.+S-1) in 5 weeks was given to patients with R(n=51) and BR-V(n=14) PDAC.
Results: There was no significant difference in adverse event rate and completion rate of NACRT protocol between short and long NACRT. The reduction rates of CA19-9 level and SUV max were both significantly higher in patients with long NACRT than those with short NACRT (64%vs30%:P=0.009 and 51%vs23%:P< 0.0001, respectively). There was no significant difference in operation time, R0 reduction rate, Evans grade, and induction and completion rates of postoperative adjuvant chemotherapy between the two groups. However, resection rate was significantly lower in long NACRT group (96% vs 85%, P=0.041) because distant metastasis was more frequently detected before surgery. There was no significant difference in OS and RFS between R patients with short and long NACRT (P=0.871 and P=0.743, respectively). In contrast, BR-V patients with long NACRT had significantly better OS and RFS than those with short NACRT (P=0.004 and P=0.022, respectively).
Conclusions: There was no significant difference between short and long NACRT in R-PDAC. Long NACRT might be more effective against BR-V PDAC.
OP04-30 A National Analysis of the Incidence and Sequelae of Pancreatogenic Diabetes Following Pancreatic Resection
Jennifer Underhill, United States

J. Underhill, J.M. Hyer, C. Aquina, J. Cloyd, M. Dillhoff, A. Manilchuk, A. Tsung, T. Pawlik, A. Ejaz
The Ohio State University Wexner Medical Center and James Comprehensive Cancer Center, United States

Introduction: New-onset diabetes following pancreatic resection (pancreatogenic diabetes mellitus, P-DM) is a known risk factor. However, the long-term incidence of P-DM and its clinical impact following pancreatic resection remains unknown.
Methods: The Medicare 100% Standard Analytic File (2013-2017) was queried for all patients who underwent partial pancreatic resection (pancreaticoduodenectomy, distal pancreatectomy). The primary outcome was the development of postoperative P-DM following surgery.
Results: We identified 4,255 patients who underwent a pancreaticoduodenectomy (n=2,989, 70.2%) or distal pancreatectomy (n=1,266, 29.8%). After a median follow-up of 0.9 years, the incidence of P-DM was 25.4% (n=863) and occurred at a median of 0.3 years following surgery. Risk factors for developing P-DM included undergoing a distal pancreatectomy (OR 1.98, 95%CI 1.67-2.34), having a malignant diagnosis (OR 1.66, 95%CI 1.35-2.05), and a family history of diabetes (OR 2.10, 95%CI 1.46-3.03) all (p< 0.001). Patients who developed P-DM were more commonly readmitted within 90 days (43% vs. 33.7%) and had higher postoperative healthcare expenditures in the year following surgery ($24,440 USD vs. $16,130 USD) (both p< 0.001) compared to patients who remained diabetes-free.
Conclusion: Approximately 1 in 4 Medicare beneficiaries who undergo a pancreatic resection develop pancreatogenic diabetes following pancreatic resection. Appropriate screening and improved patient education should be conducted for these patients, particularly those at highest risk.
OP04-31 Pre-operative Prediction of Outcome in Patients Undergoing Whipple's Pancreatoduodenectomy: Prospective Validation of a Novel Risk Scoring
Samrat Ray, India

S. Ray1, S. Das1, V. Mangla1, A. Yadav1, P. Chugh2, N. Mehta1, S. Nundy1
1Surgical Gastroenterology and Liver Transplantation, Sir Ganga Ram Hospital, India, 2Research and Biostatistics, Sir Ganga Ram Hospital, India

Introduction: Despite a high morbidity following Whipple's pancreatoduodenectomy, there is a lack of an objective pre-operative tool, based only on clinical and biochemical parameters to predict the outcome following pancreatoduodenectomy that could be implemented on an outpatient basis.
Methods: Using a multivariate regression model, the significant predictors of post-operative outcome were identified in a set of retrospective database of patients (2006-2017), and a risk score developed by binary logistic regression method. This was validated in a set of prospective patients (2017-2020). The model's predictive accuracy and discriminative ability were assessed using the receiver operating characteristics (ROC) analysis and Hosmer-Lemeshow goodness of fit tests respectively.
Results: On multivariate analysis in the retrospective cohort (n=442), the significant predictors of post-operative outcome were identified as peak bilirubin levels, pre-operative stenting and diagnosis (benign/malignant). A risk score was derived and validated on the prospective cohort (n=185) [Table 1]. The mean risk for an unfavourable outcome was 24% for a score of < /=7, 44% for a score of 8-14 and 70% for a score of >/=15. This was further tested on the validation cohort for individual risk scores (AUC=0.708) and scores categorised (AUC=0.698). There was no significant difference between observed and expected risk of major complications (p=0.31).
Conclusion: The risk score showed a fair accuracy in predicting post-operative morbidity in the prospective cohort. Therefore, we propose this be used as a quick aid to predict the operative outcome in patients posted for pancreatoduodenectomy on an outpatient basis using simple pre-operative clinical and laboratory variables.
VariablesCategoriesBeta CoefficientP valueRisk score
Peak Bilirubin1) < 2 mg/dl; 2) 2-5 mg/dl; 3) 5-10 mg/dl; 4) 10-20 mg/dl; 5) >20 mg/dl0.0060.0011) 0; 2) 1; 3) 3; 4) 7; 5) 11
Stenting1) Stented; 2) Unstented1.143<0.0011) 0; 2) 9
Diagnosis1) Benign; 2) Malignant0.661<0.0011) 0; 2) 5
Total   25 (max)
[Pre-operative risk scoring model]
OP04-32 Artery-First Pancreatectomy with Superior Mesenteric - Portal Vein Resection and Reconstruction: Two Large Institutions Experience from East and West
Atsushi Oba, Japan

A. Oba1,2, K. Tanaka3,4, E. Rangelova3, Y. Inoue1, H. Ito1, Y. Takahashi1, A. Saiura1, R. Schulick2, M. Del Chiaro2,3
1Cancer Institute Hospital, Japanese Foundation for Cancer Research, Japan, 2University of Colorado School of Medicine, United States, 3Karolinska Institutet, Sweden, 4Hokkaido University Faculty of Medicine, Japan

Introduction: Potential benefits of pancreatectomies associated to vein resection (PAVR) for pancreatic cancer are still contradictory. Although some recent papers suggested artery-first approach facilitated PAVR, evidence is sparse. The aim of this study is to analyze outcomes of artery-first approach with PAVR by using two large institutions from different regions.
Methods: We identified consecutive series of patients with pancreatic cancer who underwent artery-first approach with PAVR in Karolinska University Hospital (KUH) and Cancer institute hospital, Japanese foundation of cancer research (JFCR) from 2008 to 2018. We compared the short- and long-term results between two centers.
Results: Among total 506 patients, 211 patients were from KUH and 295 patients were from JFCR. The higher incidence of total pancreatectomy was shown in KUH (24.6% vs 0.3%, P < 0.001). The higher incidence of primary end-to-end anastomosis was shown in JFCR (92.5% vs 62.6%, P = 0.017). There was no significant difference in intraoperative estimated blood loss (KUH: 630ml, JFCR: 600ml), severe complications rate (8.5%, 5.1%), and mortality (2.4%, 0.7%). Primary end-to-end anastomosis was mainly performed even if the length of PV/SMV resection was 5cm or more and achieved successfully without thrombus (overall cases: 98.0%, 5cm or more: 93.5%)
Conclusions: We reported favorable short-term outcomes and acceptable long-term outcomes of artery-first approach with PAVR for pancreatic cancer from the two high-volume centers in the east and west. Primary end-to-end anastomosis after artery-first pancreatectomy was safe and feasible even if the length of PV/SMV resection was 5cm or more.
[The types of vein reconstruction for each length of PV/SMV resection and the rate of no thrombus]
OP04-34 Laparoscopic versus Open Pancreatoduodenectomy: An Individual Patient Data Meta-analysis of Randomized Controlled Trials
Frederique Vissers, Netherlands

F. Vissers1, J. van Hilst1,2, F. Burdio3, S. Sabnis4, M. Dijkgraaf5, S. Festen2, C. Palanivelu4, I. Poves3, M. Besselink1
1Surgery, Amsterdam UMC, University of Amsterdam, Netherlands, 2Surgery, OLVG, Netherlands, 3Surgery, Hospital del Mar, Spain, 4Surgery, Gem Hospital, India, 5Clinical Research Unit, Amsterdam UMC, University of Amsterdam, Netherlands

The first randomized controlled trials (RCT) comparing LPD to OPD have been published recently with conflicting results where observational studies show less postoperative complications. An individual patient data meta-analysis (IPDMA) may give more insight in the putative differences, including in subgroups.
A systematic literature search was performed in Pubmed, Embase and the Cochrane library. Out of 1410 studies, three RCT's comparing LPD to OPD were identified. The primary outcome was major postoperative complications (Clavien-Dindo grade ≥ III). Subgroup analyses were performed for high-risk groups including patients with a BMI of ≥25 kg/m2, either a BMI of ≥25 kg/m2 and/or a pancreatic duct < 3mm, age ≥70 years, and malignancy were performed.
Individual patient data from 224 patients included from 6 centers were collected. After LPD, major complications occurred in 33/114 (29%) patients compared to in 34/110 (31%) patients after OPD (adjusted OR 0.62; 95%CI 0.27 - 1.41, p = 0.257). No differences were seen for postoperative pancreatic fistula (adjusted OR 0.78; 95%CI 0.316 - 1.943, p = 0.599), delayed gastric emptying (adjusted OR 0.56; 95%CI 0.220 - 1.418), p=0.220) and 90-day mortality [8 (7%) vs 4 (4%)] (adjusted OR 0.15; 95% CI 0.02 - 1.26, P=0.08) after LPD vs OPD. With LPD, operative time was longer (420 vs 318 minutes, p< 0.001) and primary LOHS was shorter (mean difference -6.97 days).
This IPDMA does not show benefits nor disadvantages for LPD as compared to OPD besides a shorter LOHS. Subgroup analyses showed similar postoperative outcomes in the high-risk subgroups.
OP04-33 Impact of Borderline Resectability in Pancreatic Head Cancer on Patient Survival: Biology Matters According to the New International Consensus Criteria
Stefan Löb, Germany

F. Anger1, A. Doering1, J.-F. Lock1, C.-T. Germer1, I. Klein1, A. Wiegering1, V. Kunzmann2, C. van Eijck3, S. Löb1
1Department of General-, Visceral- and Transplantsurgery, University Hospital Wuerzburg, Germany, 2Department of Internal Medicine II, University Hospital Wuerzburg, Germany, 3Erasmus MC - University Medical Center, Netherlands

Background: International consensus criteria (ICC) have redefined borderline resectability for pancreatic ductal adenocarcinoma (PDAC) according to three dimensions: anatomical (BR-A), biological (BR-B) and conditional (BR-C). Aim of this study was to evaluate the impact of the novel consensus criteria defining BR-PDAC compared to current NCCN guidelines on patient survival after upfront pancreaticoduodenectomy.
Methods: Patients' tumours were retrospectively defined borderline resectable according to ICC. The study cohort was grouped into either BR-A or BR-B and compared to patients considered primarily resectable (R). Differences in postoperative complications, pathological reports, overall (OS) and disease-free survival (DFS) were assessed.
Results: 223 patients underwent resection for PDAC. By applying ICC in routine preoperative assessment, 20 patients were classified as stage BR-A and 36 patients as stage BR-B. 167 patients were considered resectable (R). The cohort did not contain BR-C patients. No differences in postoperative complications were detected. Median OS was significantly shorter in BR-A (12 months) and BR-B (14 months) compared to R (20 months) patients (BR-A vs. R: p=0.036 and BR-B vs R: p=0.016). CA19-9, as the determining factor of BR-B patients, turned out to be an independent prognostic risk factor for OS.
Conclusion: Preoperative staging defining surgical resectability in PDAC according to ICC is crucial for patient survival. Patients with PDAC BR-B should be considered for multimodal neoadjuvant therapy.
OP04-35 Total Pancreatectomy Risk Model for Severe Postoperative Complications Derived from 2,167 Patients Recorded in a Nationwide Clinical Database
Daisuke Hashimoto, Japan

D. Hashimoto1, M. Mizuma2, H. Kumamaru3, S. Satoi1, H. Yamaue4, M. Yamamoto5, Y. Kakeji6, M. Unno2, K. Okazaki7
1Department of Surgery, Kansai Medical University, Japan, 2Department of Surgery, Tohoku University, Japan, 3Department of Healthcare Quality Assessment, The University of Tokyo, Japan, 4Second Department of Surgery, Wakayama Medical University, Japan, 5Department of Surgery, Institute of Gastroenterology, Tokyo Women's Medical University, Japan, 6Division of Gastrointestinal Surgery, Department of Surgery, Kobe University, Japan, 7Department of Gastroenterology and Hepatology, Kansai Medical University, Japan

Background: Total pancreatectomy is required to completely clear tumors that are locally advanced or located in the center of the pancreas. However, reports describing the clinical outcomes after total pancreatectomy are rare. The aim of this retrospective observational study was to assess the clinical outcomes following total pancreatectomy using a nationwide registry and to create a risk model for severe postoperative complications.
Method: Patients who underwent total pancreatectomy from 2013-2017 and who were recorded in the Japan Society for Gastroenterological Surgery and Japanese Society of Hepato-Biliary-Pancreatic Surgery database were included. Severe complications at 30 days were defined as Clavien-Dindo grade III with reoperation or grade IV/V. We modeled the occurrence of severe complications among the patients from 2013-2016 and tested the accuracy of the model among the patients from 2017 using c-statistics and a calibration plot.
Results: We included 2167 patients undergoing total pancreatectomy. Postoperative 30-day and in-hospital mortality occurred in 1.0 per cent (22/2167) and 2.7 per cent (58/2167) of patients, respectively, and severe complications occurred in 6.0 per cent (131/2167) of patients. Factors showing a strong positive association with outcome in this risk model were the American Society of Anesthesiologists performance status and combined arterial resection. In the testing cohort, the c-statistic of the model was 0.70 (95 per cent confidence interval: 0.59-0.81).
Conclusion: Our risk model for severe postoperative complications after total pancreatectomy based on a nationwide clinical database showed good calibration and may improve the quality of pancreatic surgery.
OP04-36 Minimally Invasive versus Open Distal Pancreatectomy: An Individual Patient Data Meta-analysis of Two Randomized Controlled Trials
Maarten Korrel, Netherlands

M. Korrel1, F. Vissers1, S. Festen2, B. Groot Koerkamp3, M. Luyer4, P. Sandström5, M. Abu Hilal6, M. Besselink1, B. Björnsson5, International Minimally Invasive Pancreatic Resection Trialists Group
1Amsterdam UMC, University of Amsterdam, Netherlands, 2OLVG, Netherlands, 3Erasmus University Medical Center, Netherlands, 4Catharina Hospital Eindhoven, Netherlands, 5Linköping University Hospital, Sweden, 6Brescia University Hospital, Italy

Background: Minimally invasive distal pancreatectomy (MIDP) may reduce overall complications and hospital stay as compared to open distal pancreatectomy (ODP). This study aimed to combine data of randomized controlled trials (RCTs) comparing MIDP vs. ODP and assess treatment effects in different high-risk subgroups by conducting an individual patient data meta-analysis.
Methods: The principal investigators of the LEOPARD trial from the Netherlands and the LAPOP trial from Sweden agreed to perform this study upon completion of the trials. After completion of both trials, individual patient data will be obtained, and data collection, definitions, and outcomes harmonized. The primary endpoint is the overall rate of major (Clavien-Dindo ≥III) complications. Secondary outcomes include length of stay and individual major complications. Sensitivity analyses will be performed in three pre-specified subgroups (i.e. BMI ≥25 kg/m2, severe comorbidity and malignant disease).
Results: Results of the LAPOP trial are not yet published. These results will have been published during IHPBA 2020, and therefore, results of this individual patient data meta-analysis will become available when the congress will take place.
Conclusions: This is the first individual patient data meta-analysis including RCTs on MIDP vs. ODP, creating the largest sample of randomized patients in this field. In this study, two individual trial teams, jointly working together as the International Minimally Invasive Pancreatic Resection Trialists Group, will combine individual patient data. Results of this individual patient data meta-analysis will be presented when accepted for IHPBA 2020.
OP04-40 The Role of Radical Antegrade Modular Pancreatosplenectomy Compared to Conventional Distal Pancreatosplenectomy in Patients with Left-sided Pancreatic Cancer: A Retrospective Multicenter Propensity- Score Matching Analysis
Naru Kim, Korea, Republic of

N. Kim1, C.-S. Lim2, Y.C. Shin3, W.H. Jung4, J.S. Heo1, D.W. Choi1, S.H. Shin1, I.W. Han1
1Division of Hepatobiliary-Pancreatic Surgery, Department of Surgery, Samsung Medical Center, Sungkyunkwan University, Korea, Republic of, 2Department of Surgery, Seoul Metropolitan Government - Seoul National University Boramae Medical Center, Seoul National University College of Medicine, Korea, Republic of, 3Department of Surgery, Ilsan Paik Hospital, Inje University College of Medicine, Korea, Republic of, 4Department of Surgery, Ajou University School of Medicine, Korea, Republic of

Introduction: The purpose of this study aimed to evaluate the role of radical antegrade modular pancreatosplenectomy (RAMPS) in terms of postoperative outcomes compared to conventional distal pancreatosplenectomy (DPS) in patients with left- sided pancreatic ductal adenocarcinoma (PDAC).
Method: From 2005 to 2017, consecutive 316 left-sided PDAC patients who underwent RAMPS (n=236) or DPS (n=80) for curative intent in four tertiary referral hospitals in Korea were included in this study. Among these, after 1:2 Propensity score matching with age, sex, differentiation, T and N stage, 71 patients with DPS and 139 patients with RAMPS were analyzed for clinicopathological outcomes.
Result: There was no difference in complication rate between the two groups. RAMPS was superior than DPS in terms of R0 rate (99.3% vs 88.6%, p< 0.01) and harvested LN numbers (16.3±10.57 vs. 10.3±7.06, p< 0.01). RAMPS showed prolonged DFS (median survival 11 vs 9months), but statistically not significant (p=0.148). In a subgroup analysis with node-negative patients (n=107), RAMPS showed superior disease-free survival than DPS with statistically marginal significance (MS 15 vs 9m, p= 0.05). After multivariate analysis, preoperative CA19-9≥37, tail cancer, poorly or undifferentiated carcinoma, R1 resection, and absence of adjuvant treatment were identified as independent risk factors for survival. Also, preoperative CA19-9≥37, advanced T stage, LN metastasis, poorly or undifferentiated carcinoma were independent risk.
Conclusion: Although we could not find an eminent survival benefit of RAMPS, it could be considered a standard surgical method for left-sided PDAC because of the similar complication rate and several oncologic benefits.
OP04-42 The Yonsei Experience of Minimal Invasive Pancreaticoduodenectomies: A Propensity Score‑Matched Analysis with Open Pancreaticoduodenectomy
Munseok Choi, Korea, Republic of

M. Choi, H.K. Hwang, W.J. Lee, C.M. Kang
Department of Surgery, Yonsei University College of Medicine, Korea, Republic of

Introduction: With continued technical advances in surgical instruments and growing expertise, several surgeons have performed minimal invasive pancreaticoduodenectomy (MIPD) safely with good results, and the approach is being performed more frequently. We performed over 200 cases of MIPD and compared their outcomes to those of open pancreaticoduodenectomy (OPD) using the large sample size. The aim of the present study was to evaluate the safety and feasibility of MIPD compared with OPD.
Methods: From September 2012 to December 2019, pancreaticoduodenectomy was performed for 352 patients at Yonsei University Severance Hospital by a single surgeon. Patients were divided into two groups: those who underwent OPD (n=132) and those who underwent MIPD (n=220). We performed a 1:1 propensity score-matched analysis and retrospectively analyzed the demographic and surgical outcomes.
Results: After Propensity score matching analysis, the mean operation time for the MIPD group was similar and estimated blood loss was lower than the OPD group. The postoperative pancreatic fistula (POPF) grade B and C did not differ significantly between the 2 groups (p=0.204). There was no difference in 30-day mortality rates between the two groups (p=1.000).
Conclusions: MIPD can be a good alternative option for well-selected patients with periampullary lesions requiring pancreaticoduodenectomy.
OP04-43 The Outcome of Laparoscopic Pancreaticoduodenectomy Is Improved with the Learning Curve and Patients' Selection. Analysis in 130 Patients
Béatrice Aussilhou, France

S. Dokmak, B. Aussilhou, F.S. Ftériche, O. Soubrane, A. Sauvanet
HBP Departement and Liver Transplantation, Beaujon Hospital, France

Introduction: In our first experience Laparoscopic pancreaticoduodenectmoy (LPD) was associated with higher morbidity. Since we restrict LPD to patients at lower risk of pancreatic fistula (PF) and we ameliorate our surgical technique. We analyzed our recent results.
Methods and patients: Between 2011-2018, 130 pure LPD were performed, divided in 3 consecutive periods: period 1 (n=43), period 2 (n=43=) and period 3 (n=44) and were compared.
Results: In the third period, more females (48%, 46%, 59%, p=0.12), IPMN become the first indication of LPD (12%, 39%, 34%; p=0.037) followed by ampulloma (30%, 9%, 20%), less resection for pancreatic adenocarcinoma (35%, 16%, 16%; p=0.004), and more dilated (>3mm) wirsung duct > 3 mm (16%, 27% and 57%; p< 0.001). The third period showed less operative time (330, 345, 270; p< 0.001) and blood loss (300, 200 125; p< 0.001). Similar mortality (4%, 4%, 2%; p=0.53), decrease in all complications including mainly grades B/C PF (44%, 28%, 20%; p=0.017), bleeding (28% ,21%, 14%, p=0.26), re-intervention (19%, 14%, 9%; p=0.43) and hospital stay (26, 19, 18; p=0.045). In patients with adenocarcinoma (n=69), similar tumor size but more harvested lymph nodes (21, 19, 25; p=0.031) and R0 resection (70%, 79%, 84%; p=0.5). On multivariate analysis protective factors against grades B/C PF were female gender, pancreatic adenocarcinoma, BMI < 22.5 and the third period.
Conclusion: With patient selection and the learning curve, the results of LPD are improved. These results are important for the safe implementation of this technique.
OP04-44 Peripancreatic Bacterial Contamination Can Lead to the Development of Postoperative Pancreatic Fistula after Pancreaticoduodenectomy
Norihisa Kimura, Japan

N. Kimura, K. Ishido, T. Wakiya, H. Nagase, K. Hakamada
Gastroenterological Surgery, Hirosaki University Graduate School of Medicine, Japan

Introduction: The aim of this study was to analyze the relationship between Peripancreatic bacterial contamination (PPBC) and postoperative pancreatic fistula (POPF) after pancreaticoduodenectomy (PD) and to investigate the bacterial species in the peripancreatic fluid and identify useful antibiotics to prevent POPF.
Methods: Three hundred twenty consecutive patients underwent PD between May 2012 and December 2019. Amylase (D-AMY) and microbial culture have been routinely obtained from the peripancreatic drain on postoperative day (POD) 1, 3, and 6 since May 2012 and Modified Blumgart pancreaticojejunostomy (MBPJ) (N=158) has been adopted since May 2016. POPF was defined as grade B/C according to the international definition.
Results: POPF occurred in 26 (16.5%) of 158 patients with MBPJ. In univariate analysis, non-pancreatic disease (P=0.026), body mass index (BMI) >25 kg/m2 (P=0.016), soft pancreas (P=0.009), D-AMY on POD1 >5000 U/L (P< 0.001), and PPBC on POD1 or 3 (P< 0.001) were significantly associated with POPF. In multivariate analysis, BMI >25kg/m2 (Odds ratio [OR]=3.61; P=0.022), D-AMY on POD1 >5000 U/L (OR=5.28; P=0.004) and PPBC on POD1 or 3 (OR=4.96; P=0.003) were independent risk factors of POPF. Of all 320 patients, PPBC gradually increased from POD1 to 6. The most commonly isolated bacteria on POD1 or 3 were Enterococcus sp. (10.9%), Pseudomonas sp. (2.5%), and Enterobacter sp. (1.9%) which were sensitive to piperacillin, imipenem, meropenem, and levofloxacin.
Conclusions: Early PPBC after PD can cause the development of POPF. The patients suspected of PPBC should receive more sensitive antibiotics in the early postoperative period to prevent severe POPF.
OP04-45 Delayed Gastric Emptying in Diabetic Patients Undergoing Pancreaticoduodenctomy: A Procedure-targeted NSQIP Analysis
Alissa Greenbaum, United States

A. Greenbaum, A. Kangas-Dick, D. Moore, E. Kenny, V. Gall, D. August
Rutgers Cancer Institute of New Jersey, United States

Introduction: Delayed gastric emptying (DGE) is a major source of morbidity after pancreaticoduodenctomy (PD). Patients with diabetes mellitus (DM) have a propensity for gastric dysmotility, however the relationship between DGE and DM is not clearly established. The aim of this study was to determine the incidence of DGE in patients with and without DM after PD.
Methods: The American College of Surgeons National Quality Improvement Project procedure-targeted pancreatectomy database was queried from 2014-2017 for patients undergoing PD and combined with the main database. Variables were compared by DM status. The primary outcome was DGE.
Results: 14,735 patients met inclusion criteria, including 10,930 non-DM (74.2%) and 3805 DM patients (25.8%). DGE occurred in 17.1% (n=2519); 17.2% in non-DM and 16.8% in DM patients (p=0.60). DM patients had increased rates of hypertension and pancreatic adenocarcinoma, larger duct size, harder gland texture, and required more vascular resections (p< 0.001). DM patients had equivalent rates of postoperative infections compared to non-DM patients. IDDM had better outcomes compared to non-IDDM patients in organ space infections (8.8 vs 14.2%;p< 0.001) and pancreatic fistula (11.7 vs 18.3%;p< 0.001). Rates of DGE were 18.0% (n=337) in non-IDDM and 15.7% in IDDM patients (p=0.162). On multivariate regression, male sex, advanced age, smoking, pancreatic fistula, and organ space infection were associated with DGE.
Conclusion: No differences in rates of DGE between DM and non-DM patients after PD were found. IDDM patients demonstrated better postsurgical outcomes compared to non-IDDM patients, suggesting a potential role of monitored insulin or hyperglycemia regulation.
OP04-46 Laparoscopic Central Pancreatectomy: Results in 81 Patients
Béatrice Aussilhou, France

S. Dokmak, B. Aussilhou, F.S. Ftériche, O. Soubrane, A. Sauvanet
HBP Departement and Liver Transplantation, Beaujon Hospital, France

Introduction: Central pancreatectomy (CP) is a good indication to the laparoscopic approach related to the absence of oncological or vascular contraindications. The aim of this study was to analyze our monocentric experience.
Methods: Between 2008-2018, were performed 540 laparoscopic pancreatic resections and 81 laparoscopic CP. CP was indicated if enucleation was not feasible, in non-diabetic patients and if the distal pancreas was > 5 cm. One layer pancreato-gastric anastomosis. All clinical, operative and postoperative data were recorded prospectively and were analyzed.
Results: The mean age was 50 (17-77), including 55 female (68%), with a mean BMI at 25 (16-36). Indications for resection were for neuroendocrine tumor (24; 30%), IPMN (16; 20%), solid pseudopapillary tumor (12; 15%), mucinous cystadenoma (11; 14%), pancreatitis with disconnected duct syndrome (5; 6%), and other (13; 15%). The mean operative time was 183 (90-285), the mean blood loss 107 (0-800), and one conversion (1%). No 90 days mortality and the overall morbidity was observed in 58 patients (72%) including grade B/C pancreatic fistula (21; 26%), bleeding (10; 12%), drained collection (2; 3%), delayed gastric emptying (2; 3%), re-intervention (5; 6%). the mean hospital stay was 22 days (5-54) with readmission in 2 (2%). The mean number of harvested lymph nodes was 3 (0-19) and R0 resection in 71 (88%) patients.
Conclusion: The applicability of laparoscopic central pancreatectomy is high and the morbidity is acceptable. There is a real advantage on the preservation of the pancreatic function and abdominal wall in these young patients with no malignancy.
OP04-48 Surgical Complications after Preoperative Chemoradiotherapy in Patients with Resectable and Borderline Resectable Pancreatic Cancer in a Multicentre, Randomised Controlled Clinical Trial (PREOPANC-1)
Jelle Corneel van Dongen, Netherlands

J.C. van Dongen1, E. Versteijne2, M. Suker1, B.A. Bonsing3, M.G. Besselink2, O.R. Busch2, G. van Tienhoven2, B. Groot Koerkamp1, C.H. van Eijck1, Dutch Pancreatic Cancer Group
1Erasmus MC - University Medical Center, Netherlands, 2Amsterdam UMC, University of Amsterdam, Netherlands, 3Department of Surgery, Leiden University Medical Center, Netherlands

Background: Preoperative chemoradiotherapy is increasingly being used in patients with (borderline-)resectable pancreatic cancer. However, randomised studies investigating the effect of preoperative therapy on the surgical complication rate after pancreatic resection are lacking.
Objectives: To investigate the effect of preoperative chemoradiotherapy on surgical complications in patients after pancreatic resection for (borderline-)resectable pancreatic cancer.
Methods: In this prospective, multicentre, randomised controlled trial, patients with (borderline-)resectable pancreatic cancer were randomly assigned (1:1) to upfront surgery, followed by adjuvant therapy or to preoperative chemoradiotherapy followed by surgery and adjuvant chemotherapy. The endpoints of our study were the rate of postoperative pancreatic fistula (POPF), post pancreatectomy haemorrhage (PPH), delayed gastric emptying (DGE), bile leakage, intra-abdominal infections, major complications and mortality.
Results: This study included 218 patients, of which 84 underwent curative resection in the upfront surgery group (75%) and 60 in the preoperative therapy group (57.1%). There was a higher incidence of POPF in the group who underwent upfront surgery compared to preoperative chemoradiotherapy (10.7% vs. 0%, p = 0.011). The incidence of PPH did not differ significantly between the two treatment groups (7.6% vs. 10.7%, p=0.553, respectively), but a different etiology was observed. The upfront surgery group included five (6.0%) patients with late extra-luminal PPH, compared to zero in the preoperative chemoradiotherapy group (p=0.076). No significant differences were found regarding other surgical complications.
Conclusion: Preoperative chemoradiotherapy does not increase the incidence of surgical complications or mortality. In contrast, it was associated with a reduced POPF rate.