Pancreatic P-P01 Impact of gastric resection and enteric anastomotic configuration on delayed gastric emptying after pancreaticoduodenectomy: a network meta-analysis Chris Varghese 1 , Sameer Bhat 1 , Tim Hsu Wang 1 , Khaled Ammar 2 , Greg O’Grady 1 , Sanjay Pandanaboyana 3 1 University of Auckland, Auckland, New Zealand, 2 Freeman Hospital, Newcastle, United Kingdom, 3 Newcastle University, Newcastle, United Kingdom Background: Delayed gastric emptying (DGE) is frequent after pancrea- ticoduodenectomy (PD). Several randomised controlled trials (RCTs) have explored operative strategies to minimise DGE, however, the opti- mal combination of gastric resection approach, anastomotic route, and configuration, role of Braun enteroenterostomy remains unclear. Methods: MEDLINE, Embase, and CENTRAL databases were systemati- cally searched for RCTs comparing gastric resection (Classic Whipple, pylorus-resecting, and pylorus-preserving), anastomotic route (ante- colic vs retrocolic) and configuration (Billroth II vs Roux-en-Y), and enteroenterostomy (Braun vs no Braun). A random-effects, Bayesian network meta-analysis with non-informative priors was conducted to determine the optimal combination of approaches to PD for minimis- ing DGE. Results: Twenty-four RCTs, including 2526 patients and 14 approaches were included. There was some heterogeneity, although inconsistency was low. The overall incidence of DGE was 25.6% (n ¼ 647). Pylorus- resecting, antecolic, Billroth II with Braun enteroenterostomy was associated with the lowest rates of DGE and ranked the best in 35% of comparisons. Classic Whipple, retrocolic, Billroth II with Braun ranked the worst for DGE in 32% of comparisons. Pairwise meta-analysis of ret- rocolic vs antecolic route of gastro-jejunostomy found increased risk of DGE with the retrocolic route (OR 2.1, 95% CrI; 0.92 - 4.7). Pairwise meta- analysis of Braun enteroenterostomy found a trend towards lower DGE rates with Braun compared to no Braun (OR 1.9, 95% CrI; 0.92 - 3.9). Having a Braun enteroenterostomy ranked the best in 96% of compari- sons. Conclusions: Based on existing RCT evidence, a pylorus-resecting, antecolic, Billroth II with Braun enteroenterostomy may be associated with the lowest rates of DGE. P-P02 Incidence and risk factors for chyle leak after pancreatic surgery: a systematic review and meta-analysis Chris Varghese 1 , Cameron Wells 1 , Shiela Lee 2 , Khaled Ammar 3 , Sanjay Pandanaboyana 2 1 University of Auckland, Auckland, New Zealand, 2 Freeman Hospital, Newcastle, United Kingdom, 3 Free, Newcastle, United Kingdom Background: The incidence of, and risk factors for chyle leak, as de- fined by the 2017 International Study Group on Pancreatic Surgery (ISGPS), remain unknown. Methods: MEDLINE, EMBASE, and Scopus were systematically searched for studies of patients undergoing pancreatectomy that reported chyle leak according to the 2017 ISGPS definition. The primary outcomes were the incidence of overall and clinically-relevant chyle leak. A random-effects pairwise meta-analysis was used to identify risk fac- tors where possible. Results: Thirty-five studies including 7083 patients were included in the meta-analysis. The weighted incidence of overall chyle leak was 6.8% (95% CI 5.6 - 8.2) and clinically-relevant chyle leak was 5.5% (95% CI 3.8 - 7.7). Pancreaticoduodenectomy, total pancreatectomy and distal pancreatectomy were associated with a CL incidence of 7.3%, 4.3%, 5.8% respectively. Fourteen individual risk factors for chyle leak were identified from included studies. Younger age, low prognostic nutri- tional index, para-aortic node manipulation, lymphatic involvement, and post-pancreatectomy pancreatitis were significantly associated with chyle leak, all from individual studies. Conclusions: The incidence of overall chyle leak and clinically relevant chyle leak after pancreatic surgery, as defined by the 2017 ISGPS defini- tion is 6.8% and 5.5% respectively. Several risk factors for chyle leak were identified in the present review, however, larger high-quality studies are needed to more accurately define these risks. P-P03 Impact of routine nasogastric decompression versus no nasogastric decompression after pancreatoduodenectomy on perioperative outcomes: A meta-analysis of available evidence Khaled Ammar 1 , Chris Varghese 2 , Thejasvin K 3 , Viswakumar Prabakaran 4 , Stuart Robinson 4 , Samir Pathak 5 , Bobby VM Dasari 6 , Sanjay Pandanaboyana 4 1 HPB and Transplant surgery department, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 2 Department of Surgery, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand, 3 Department of Surgery, Faculty of Medical Sciences, Newcastle upon Tyne, United Kingdom, 4 HPB and Transplant surgery, Freeman Hospital, Newcastle upon Tyne, United Kingdom, 5 HPB surgery, St James Hospital, Leeds, United Kingdom, 6 HPB and Transplant surgery, Queen Elizabeth Hosptial, Birmingham, United Kingdom Background: This meta-analysis reviewed the current evidence on the impact of routine Nasogastric decompression (NGD) versus no NGD af- ter pancreatoduodenectomy on perioperative outcomes. Methods: PubMed, Medline, Scopus, Embase and Cochrane databases were searched for studies reporting the role of nasogastric tube decom- pression after pancreatoduodenectomy on perioperative outcomes were retrieved and analysed up to January 2021. Results: Eight studies with total of 1301 patients were enrolled of which 668 patients had routine NGD. Routine NGD was associated with a higher incidence of overall delayed gastric emptying (DGE) and clini- cally relevant DGE (OR ¼ 2.51, 95% CI; 1.12 - 5.63, I2¼ 83%, P ¼ 0.03, and OR ¼ 3.64, 95% CI: 1.83 – 7.25, I2 ¼ 54%, P < 0.01, respectively). Routine NGD was also associated with a higher rate of Clavien-Dindo  2 com- plications (OR ¼ 3.12, 95% CI: 1.05 – 9.28, I2 ¼ 88%, P ¼ 0.04), and in- creased length of hospital stay (MD ¼ 2.67, 95% CI: 0.60 – 4.75, I2 ¼ 97%, P ¼ 0.02). There were no significant differences in overall complications (OR ¼ 1.07, 95% CI: 0.79 – 1.46, I2 0%, P ¼ 0.66), or postoperative pancre- atic fistula (OR ¼ 1.21, 95% CI: 0.86 – 1.72, I2 ¼ 0%, P ¼ 0.28) between the two groups. Conclusions: Routine NGD may be associated with increased rates of DGE, major complications and longer length of stay after pancreato- duodenectomy. P-P04 The cost of acute pancreatitis is amylasing! Harry Carr, Timothy Morris, Matthew Williams, Georgina Jacob, Michael Courtney, Venkatesh Kanakala Department of General Surgery, James Cook University Hospital, Middlesbrough, United Kingdom Background: Amylase is the key serum biomarker in the diagnosis of acute pancreatitis, however there is no indication for repeat/serial measurement once the diagnosis is established. It is estimated that £27,000pa is spent unnecessarily on repeat amylase investigations without clinical indication 1 . Anecdotally, within the department, unnecessary repeats were being routinely performed. Resultantly, we audited in 2019 and 2020 to un- derstand the extent of the issue. Following the first cycle, pre-rotation departmental talks were given to all relevant healthcare staff. Aim(s): • Determine and understand the extent of repeat amylase investiga- tions and promoting factors. • Assess the use of imaging. Methods: Data was collected and analysed retrospectively over 2 audit cycles (C1 & C2) from 79 patient episodes of admissions to the surgical department of a Northern Major Trauma Centre with confirmed diag- noses of acute pancreatitis between 01/05/2019 – 31/07/2019 and 01/08/ 2020 – 31/12/2020. Resources used included: patient notes, IMPAX and WebICE. Data was collected and analysed by one author in C1 but multiple authors in C2. Results: Mean age ¼ 60 years. Male:Female ratio was 8:16 and 24:31, respectively. Initial amylase was diagnostic in > 75% (61/79). 81 unnecessary repeats performed. Abstracts | ix95 Downloaded from https://academic.oup.com/bjs/article/108/Supplement_9/znab430.228/6462749 by guest on 25 January 2024