SAGES SAFE CHOLE program changes surgeons practice in France-results of the FCVD implementation of SAFE CHOLE in France.

BACKGROUND: With the Society of Gastrointestinal and Endoscopic Surgeons supervision, the Safe Cholecystectomy Task Force (SAFE CHOLE) was translated into French by the the Federation of Visceral and Digestive Surgery (FCVD) and adopted to run on its national e-learning platform for surgical continuing medical education (CME). The objective of this study was to assess the impact of the SAFE CHOLE (SF) program on the knowledge and practice of French surgeons performing cholecystectomy and participating in the FCVD lead CME activity. METHODS: To obtain CME certification, each participant must fill out three FCVD validated questionnaires regarding (1) the participants' routine practice for cholecystectomy, (2) the participants' knowledge and practice after successful completion of the program, and (3) the educational value of the SC program. RESULTS: From 2021 to 2022, 481 surgeons completed the program. The overall satisfaction rate for the program was 81%, and 53% of the surgeons were practicing routine cholangiography before the SC program. Eighty percent declared having acquired new knowledge. Fifty-six percent reported a change in their practice of cholecystectomy. Of those, 46% started routinely using the critical view of safety, 12% used a time-out prior transection of vital structures, and 11% adopted routine intraoperative cholangiography. Sixty-seven percent reported performing a sub-total cholecystectomy in case the CVS was unobtainable. If faced with BDI, 45% would transfer to a higher level of care, 33% would seek help from a colleague, and 10% would proceed with a repair. Ninety percent recommended adoption of SC by all general surgeons and 98% reported improvement of patient safety. CONCLUSIONS: Large-scale implementation of the SC program in France is feasible within a broad group of diverse specialty surgeons and appears to have a significant impact on their practice. These data should encourage other surgeons and health systems to engage in this program.

An Unmodulated Very-Low-Voltage Electrosurgical Technology Creates Predictable and Ultimate Tissue Coagulation: From Experimental Data to Clinical Use.

Objective. We analyzed the underlying principles of an unmodulated very-low-voltage (VLV) mode, designated as "soft coagulation" in hemostasis, and demonstrate its clinical applications. Summary Background Data. While the advantage of the VLV mode has been reported across surgical specialties, the basic principle has not been well described and remains ambiguous. Methods. Characteristics of major electrosurgical modes were measured in different settings. For the VLV mode, the tissue effect and electrical parameters were assessed in simulated environments. Results. The VLV mode achieved tissue coagulation with the lowest voltage compared with the other modes in any settings. With increasing impedance, the voltage of the VLV mode stayed very low at under 200 V compared with other modes. The VLV mode constantly produced effective tissue coagulation without carbonization. We have demonstrated the clinical applications of the method. Conclusions. The voltage of the VLV mode consistently stays under 200 V, resulting in tissue coagulation with minimal vaporization or carbonization. Therefore, the VLV mode produces more predictable tissue coagulation and minimizes undesirable collateral thermal tissue effects, enabling nerve- and function-preserving surgery. The use of VLV mode through better understanding of minimally invasive way of using electrosurgery may lead to better surgical outcomes.

Public reporting and transparency: a primer on public outcomes reporting.

INTRODUCTION: Healthcare consumers seeking accurate information about where to find quality surgical care face a confusing constellation of rating systems that lack transparency or consistency of opinion. For example, a 2016 report in Health Affairs demonstrated that no hospital was rated as a high performer by all four prominent national ratings systems: Consumer Reports, Leapfrog, Healthgrades and U.S. News & World Report (Austin et al. Health Aff 34:423-430, 2015). Surgeons should have an understanding of the current state of public reporting of quality; hospital ratings and data sources; physician ratings and data sources; and transparency of reporting. METHODS: We conducted a non-systematic review of the literature. RESULTS: Hospital quality ratings remain nebulous and there is not universal opinion on the utility of voluntary participation in ranking systems, leaving the current systems largely opinion-based. Early attempts at physician ranking systems are rudimentary at best and suffer from methodological concerns. Publicly reported metrics should be easily understandable, accessible, clinically relevant, reliable, non-punitive, and shielded from legal discovery. Transparency is increasing within institutions to help align staff to institutional objectives, while specialty specific registries are helping to standardize care pathways and outcomes measures across organizations. Measuring surgical outcomes beyond 30-day morbidity and mortality has been plagued by a lack of understanding on how to create metrics that matter; the four attributes of relevance, scientific soundness, feasibility and comprehensiveness set a high bar for the development of effective and efficient quality measures in surgery. DISCUSSION: SAGES, via the Quality, Outcomes, and Safety Committee, is committed to learning how to develop meaningful quality metrics in general surgery and will continue to work in other areas that impact quality, such as opioid prescribing, and surgeon wellness.

Estimating the Incidence of Stray Energy Burns during Laparoscopic Surgery based on Two Statewide Databases and Retrospective Rates: An Opportunity to Improve Patient Safety.

BACKGROUND: The growth of laparoscopic surgery has increased the use of laparoscopic electrosurgical devices based on radiofrequency current. Despite an improvement in most post-operative outcomes, the use of these devices can be associated with inadvertent thermal or mechanical injuries, also called accidental punctures and lacerations (APLs). APLs can occur through either operator error or system error, including insulation failure or capacitive coupling resulting in stray energy burns. Our aim was to estimate the incidence and-as a result-the impact of laparoscopic APLs. METHODS: A retrospective analysis of the Healthcare Cost and Utilization Project (HCUP) State Inpatient Database (SID) was performed for 2009 in California (CA) and Florida (FL). ICD-9 codes and current procedural terminology were used to query for five common general surgery procedures: appendectomy, cholecystectomy, fundoplication, gastric bypass, and gastroplasty with these procedures cross-referenced for any secondary procedure at the time of the initial surgery indicative of APLs. The c2 test was used for comparisons where appropriate. RESULTS: Overall, 192,794 primary laparoscopic procedures were identified in the HCUP database in CA and FL in 2009, with a similar procedure frequency distribution between CA and FL. Six hundred ninety-four procedures were complicated by APL. Gastric bypass and fundoplication were more commonly associated with APLs. CONCLUSION: In this retrospective analysis of procedures performed in CA and FL, the estimated incidence of APL was 3.6 per 1000 cases. Patient morbidity and mortality were likely related to both pilot-error injuries and stray energy burns during laparoscopy. Possible solutions to reduce surgical complications from APL include educational programs to reduce pilot error and the incorporation of fail-safe technologies to eliminate stray energy burns, such as active electrode monitoring and use of non-radiofrequency current (true cautery).

Incidental Gallbladder Carcinoma Discovered after Laparoscopic Cholecystectomy: Identifying Patients Who will Benefit from Reoperation.

BACKGROUND: Despite an early radical reoperation, recurrence and poor survival are observed in up to 40% of patients with an incidentally discovered gallbladder carcinoma (I-GBC) after undergoing a laparoscopic cholecystectomy (LC). This study seeks to identify prognostic factors after re-I-GBC resection. METHODS: A retrospective review of a prospectively maintained patient database with patients who were undergoing resection for I-GBC from January 1995 to March 2017 was performed. Prognostic factors for survival were assessed by multivariate Cox analysis. RESULTS: There were 50 consecutive patients (median age 64 years; range 38-82) undergoing reoperation 45 ± 30 days after LC. Re-resection entailed a major hepatectomy in five patients (10%) and lymphadenectomy in all patients. Ninety-day morbidity and mortality were 22 and 2%, respectively. Lymph node (LN) involvement was present in 24 (48%) patients with a mean of 5.79 ± 14.4 LN+. Median overall survival was 40 months with 1-, 3-, 5- and 10-year survival rates of 80, 50, 41 and 36%, respectively. Independent risk factors for overall survival were T3 tumours (HR = 7.58; 95% confidence intervals (CI), 2.41-23.83.) and LN involvement (HR = 3.66; 95% CI, 1.42-9.45). Patients presenting with zero, one and two risk factors had 3-year survival rates of 85, 31 and 0%, respectively, and median overall survival of 80, 22 and 13 months, respectively (p < 0.0001). CONCLUSIONS: After I-GBC discovery following an LC, T3 tumours and tumours with LN+ are characterised by poor prognosis. The presence and the identification of these prognostic factors help identify patients in need of alternative perioperative treatments.

Prognostic value of venous invasion in resected T3 pancreatic adenocarcinoma: Depth of invasion matters.

BACKGROUND: Incomplete evaluation of venous invasion has led to conflicting results regarding the prognosis of patients undergoing pancreatectomy with a synchronous venous resection. This study evaluates the prognostic value associated with the presence and the depth of venous invasion in T3 pancreatic adenocarcinoma. METHODS: This study evaluated retrospectively 181 consecutive pancreatoduodenectomies performed for T3N0M0 and T3N1M0 pancreatic adenocarcinomas (stages IIA and IIB) from January 2006 to December 2014. Univariate and multivariate Cox analyses were performed to assess survival prognostic factors. RESULTS: Pancreatoduodenectomies with a segmental venous resection was performed on 91 patients, while 90 other patients had a standard pancreatoduodenectomies without venous resection. Pathologic venous invasion was detected in 68 (74%) of the 91 venous resection patients. Depth of venous invasion was into the adventitia (n = 25), media (n = 28), and intima (n = 15). The overall survival rates at 1, 3, 5, and 10 years were 75%, 33%, 21%, and 6%, respectively. There were no differences in survival between patients undergoing standard pancreatoduodenectomies and pancreatoduodenectomies with venous resection (27 vs 22 months; P = .28) or between patients with and without venous invasion (20 vs 27 months; P = .08). In multivariate analysis, depth of venous invasion into the intima (hazard ratio, 2.25; 95% confidence interval, 1.16-4.34; P = .0001) and adjuvant chemotherapy (hazard ratio, 0.16; 95% confidence interval, 0.09-0.43; P ≤ .0001) were identified as independent prognostic factors of overall survival. CONCLUSION: Depth of venous invasion into the intima indicates poor survival in pancreatic T3 adenocarcinoma. Preoperative identification of this factor could be helpful for better selection of patients for curative operation.

Fundamental Use of Surgical Energy (FUSE): An Essential Educational Program for Operating Room Safety.

Operating room (OR) safety has become a major concern in patient safety since the 1990s. Improvement of team communication and behavior is a popular target for safety programming at the institutional level. Despite these efforts, essential safety gaps remain in the OR and procedure rooms. A prime example is the use of energy-based devices in ORs and procedural areas. The lack of fundamental understanding of energy device function, design, and application contributes to avoidable injury and harm at a rate of approximately 1 to 2 per 1000 patients in the US. Hundreds of OR fires occur each year in the US, some causing severe injury and even death. Most of these fires are associated with the use of energy-based surgical devices.In response to this safety issue, the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) developed the Fundamental Use of Surgical Energy (FUSE) program. This program includes a standardized curriculum targeted to surgeons, other physicians, and allied health care professionals and a psychometrically designed and validated certification test. A successful FUSE certification documents acquisition of the basic knowledge needed to safely use energy-based devices in the OR. By design FUSE fills a void in the curriculum and competency assessment for surgeons and other procedural specialists in the use of energy-based devices in patients.

Quantifying inadvertent thermal bowel injury from the monopolar instrument.

BACKGROUND: Insulation defects are observed in 3-39 % of laparoscopic instruments. Electrosurgical injuries due to insulation defects or capacitive coupling remain an issue in laparoscopic surgery with a prevalence of 0.6-5 per thousand cases. Shielded instruments with active electrode monitoring (AEM) have been postulated to prevent these injuries. The benefit of these instruments has not been quantified. Most bowel injuries are unrecognized intra-operatively. Injury is revealed only after the patient exhibits peritonitis symptoms and surgical intervention to repair the bowel is required. These injuries may result in devastating and costly complications or mortality. The extent of bowel injury possible with commonly used generator settings and associated energy output has never been histologically defined. Our objectives in this experimental study were: quantify and compare the energy released through insulation defects or capacitive coupling with standard unshielded monopolar versus shielded instruments with (AEM), determine energy required to cause a visible burn, and relate the histological burn depth to a given amount of energy. METHODS: Ex vivo porcine jejunum was used for tissue testing. An oscilloscope measured energy output from three common electrosurgical generators at recommended power settings with standard or AEM instruments with insulation defects and in capacitive coupling scenarios. Presence of a visible burn was noted, and depth of tissue damage for a given amount of energy was measured histologically. RESULTS: All samples that received ≥3.8 J of energy had visible burns. As little as 10 J caused full wall thickness burns. 3.8 J was exceeded at the 30- and 50-W power settings in every experimental scenario using standard monopolar instruments; AEM instruments never approached this much energy. CONCLUSIONS: Serious burn injury results from small amounts of energy leaked from standard instruments. AEM instruments appeared protective and did not leak sufficient energy to cause burn injuries to the bowel.

Outcomes of liver resection for haemorrhagic hepatocellular adenoma.

BACKGROUND: Intratumoral bleeding and/or intraperitoneal rupture occurs in up to 20% of patients with hepatocellular adenoma (HCA). Hepatectomy in the presence of haemorrhagic HCA has been associated with increased morbidity and mortality rates. This study evaluates the outcomes of hepatectomy for haemorrhagic HCA at a single institution. METHODS: Between January 1997 and December 2012, 52 consecutive patients underwent liver resection for HCA. Among them, 14 patients were resected for haemorrhagic (H)-HCAs (including 9 cases of intratumoural bleeding and 5 cases of intraperitoneal bleeding) and 38 for non-haemorrhagic (NH)-HCAs. RESULTS: The preoperative characteristics were similar between the two groups except for younger age (p = .001) and shorter duration of hormonal use (p = .001) in (H)-HCAs. There were no mortalities. Intraoperative blood loss, transfusion rate, and postoperative morbidity were comparable between the two groups of patients (p = ns). The length of hospital stay was significantly longer in (H)-HCAs (p = .03). In all the resected H-HCAs, pathology showed central haemorrhagic changes with tumoral cells at the periphery of the lesions. CONCLUSIONS: Liver resection for H- and NH-HCAs can be achieved with no mortality and comparable short-term outcomes.

Fundamental Use of Surgical Energy (FUSE) certification: validation and predictors of success.

INTRODUCTION: The Fundamental Use of Surgical Energy (FUSE) program includes a Web-based didactic curriculum and a high-stakes multiple-choice question examination with the goal to provide certification of knowledge on the safe use of surgical energy-based devices. The purpose of this study was (1) to set a passing score through a psychometrically sound process and (2) to determine what pretest factors predicted passing the FUSE examination. METHODS: Beta-testing of multiple-choice questions on 62 topics of importance to the safe use of surgical energy-based devices was performed. Eligible test takers were physicians with a minimum of 1 year of surgical training who were recruited by FUSE task force members. A pretest survey collected baseline information. RESULTS: A total of 227 individuals completed the FUSE beta-test, and 208 completed the pretest survey. The passing/cut score for the first test form of the FUSE multiple-choice examination was determined using the modified Angoff methodology and for the second test form was determined using a linear equating methodology. The overall passing rate across the two examination forms was 81.5%. Self-reported time studying the FUSE Web-based curriculum for a minimum of >2 h was associated with a passing examination score (p < 0.001). Performance was not different based on increased years of surgical practice (p = 0.363), self-reported expertise on one or more types of energy-based devices (p = 0.683), participation in the FUSE postgraduate course (p = 0.426), or having reviewed the FUSE manual (p = 0.428). Logistic regression found that studying the FUSE didactics for >2 h predicted a passing score (OR 3.61; 95% CI 1.44-9.05; p = 0.006) independent of the other baseline characteristics recorded. CONCLUSION(S): The development of the FUSE examination, including the passing score, followed a psychometrically sound process. Self-reported time studying the FUSE curriculum predicted a passing score independent of other pretest characteristics such as years in practice and self-reported expertise.

Long-term knowledge retention following simulation-based training for electrosurgical safety: 1-year follow-up of a randomized controlled trial.

BACKGROUND: Despite the value of simulation for surgical training, it is unclear whether acquired competencies persist long term. A prior randomized trial showed that structured simulation improves knowledge of the safe use of electrosurgery (ES) amongst trainees up to 3 months after the curriculum (Madani et al. in Surg Endosc 28(10):2772-2782, 2014). We now analyse long-term knowledge retention. This study estimates the effects of a structured simulation-based curriculum to teach the safe use of ES on knowledge after 1 year. METHODS: Trainees previously participated in a 1-h didactic ES course, followed by randomization into one of two groups: an unstructured hands-on session where trainees used ES devices (control group) or a goal-directed hands-on training session (Sim group). Knowledge of pre- and post-curriculum (immediate, 3 months and 1 year) and knowledge of ES safety were assessed using different multiple-choice examinations. Data are expressed as median (interquartile range), *p < 0.05. RESULTS: Fifty-nine trainees participated (30 control group; 29 Sim group). Despite equal baseline examination scores, Sim group demonstrated higher scores compared to control immediately (89% [83; 94] vs. 83% [71; 86]*), 3 months (77% [69; 90] vs. 60% [51; 80]*) and 1 year after curriculum (70% [61; 74] vs. 60% [31; 71]*). One-year score remained significantly greater compared to baseline in the Sim group (70% [61; 74] vs. 49% [43; 57]*), but was similar to baseline in the control group (60% [31; 71] vs. 45% [34; 52]). CONCLUSIONS: After ES simulation training, retention of competencies persists longer when the hands-on component is designed to reinforce specific learning objectives in a structured curriculum. Despite routine clinical use of ES devices, knowledge degrades overtime, suggesting the need for ongoing formal educational activities to reinforce curricular objectives.

Resection of Borderline Resectable and Locally Advanced Pancreatic Adenocarcinomas after Neoadjuvant Chemotherapy.

OBJECTIVE: To report the outcomes of surgical resection of borderline resectable (BL) and locally advanced (LA) 'unresectable' pancreatic cancer after neoadjuvant chemotherapy. METHODS: A review of a prospectively maintained database for pancreatic resections was undertaken to identify patients undergoing resection for BL and LA pancreatic cancer after neoadjuvant chemotherapy between January 2007 and December 2012. Clinicopathological, surgical and survival outcomes were analyzed. RESULTS: A total of 45 patients with LA (n = 34) or BL cancer (n = 11) underwent surgery after a mean (± SD) of 7 ± 4 preoperative chemotherapy cycles. Ninety-day mortality was 6.7%, and overall morbidity was 33.3%. An R0 resection was achieved in 34 patients, and 4 patients showed a complete pathological response. Overall median postoperative survival was 17 months (21 after the start of neoadjuvant treatment). Overall and disease-free survival was 74.9 and 43.6% at 1 year and 21.2 and 10.3% at 3 years, respectively. In BL cancer patients, the 3-year survival was significantly higher compared to that of LA cancer patients (p = 0.02). CONCLUSIONS: Curative intent resection in BL and LA cancer patients after neoadjuvant chemotherapy can be achieved with reasonable mortality and morbidity and an encouraging 3-year survival. After neoadjuvant therapy, resection provides a better overall survival for BL compared to LA cancer patients.

Predicting early intrahepatic recurrence after curative resection of colorectal liver metastases with molecular markers.

BACKGROUND: The aim of this case-control study was to identify clinicopathological factors and test three relevant biomarkers for their ability to predict early intrahepatic recurrence after curative liver resection for colorectal liver metastases (CLM). METHODS: Of the 184 patients with CLM undergoing hepatectomy between January 2007 and December 2009, thirty patients had intrahepatic disease recurrence within 6 months. The control group was randomly selected from a cohort of patients between April 1997 and December 2005 who have survived without disease recurrence after CLM resection for over 5 years. Both groups were matched for size of metastasis greater than 5.0 cm, the presence of multiple metastases, and synchronous versus metachronous CLM. The final study population consisted of 60 patients with CLM undergoing R0 hepatectomy, 30 of whom had early intrahepatic-only recurrences (study group) and 30 patients without recurrence for more than 5 years (control group). Both groups were analyzed and compared for the presence of clinical factors and expression levels of CD133, survivin, and Bcl-2 within tumor tissue. RESULTS: Characteristics of patients were similar between the two groups except primary tumor location and administration of postoperative chemotherapy. Expression level of CD133 and survivin were significantly increased in tumors of patients with recurrence compared to patients without recurrence. On multivariate analysis high tumor expression levels of CD133 (odds ratio [OR] 14.7, confidence interval [CI] 1.8-121.3, p = 0.012) and survivin (OR 9.5, CI 2.1-44.3, p = 0.004) and postoperative chemotherapy (OR 4.8, CI 1.01-22.9, p = 0.049) were independent factors associated with early intrahepatic recurrence. CONCLUSIONS: Tumor expression levels of CD133 and survivin may be a useful predictor of early intrahepatic recurrence after hepatectomy for CLM. Administration of postoperative chemotherapy may prevent early intrahepatic recurrence.

Impact of a hands-on component on learning in the Fundamental Use of Surgical Energy™ (FUSE) curriculum: a randomized-controlled trial in surgical trainees.

BACKGROUND: While energy devices are ubiquitous in the operating room, they remain poorly understood and can result in significant complications. The purpose of this study was to estimate the extent to which adding a novel bench-top component improves learning of SAGES' Fundamental Use of Surgical Energy™ (FUSE) electrosurgery curriculum among surgical trainees. METHODS: Surgical residents participated in a 1-h didactic electrosurgery (ES) course, based on the FUSE curriculum. They were then randomized to one of two groups: an unstructured hands-on session where trainees used ES devices (control group) or a goal-directed hands-on training session (Sim group). Pre- and post-curriculum (immediate and at 3 months) assessments included knowledge of ES (multiple-choice examination), self-perceived competence for each of the 35 course objectives (questionnaire), and self-perceived comfort with performance of seven tasks related to safe use of ES. Data expressed as median[interquartile range], *p < 0.05. RESULTS: 56 (29 control; 27 Sim) surgical trainees completed the curriculum and assessments. Baseline characteristics, including pre-curriculum exam and questionnaire scores, were similar. Total score on the exam improved from 46%[40;54] to 84%[77;91]* for the entire cohort, with higher immediate post-curriculum scores in the Sim group compared to controls (89%[83;94] vs. 83%[71;86]*). At 3 months, performance on the exam declined in both groups, but remained higher in the Sim group (77%[69;90] vs 60%[51;80]*). Participants in both groups reported feeling greater comfort and competence post-curriculum (immediate and at 3 months) compared to baseline. This improvement was greater in the Sim group with a higher proportion feeling "Very Comfortable" or "Fully Competent" (Sim: 3/7 tasks and 28/35 objectives; control: 0/7 tasks and 10/35 objectives). CONCLUSIONS: A FUSE-based curriculum improved surgical trainees' knowledge and comfort in the safe use of electrosurgical devices. The addition of a structured interactive bench-top simulation component further improved learning and retention at 3 months.

Use of a temporary intraoperative mesentericoportal shunt for pancreatic resection for locally advanced pancreatic cancer with portal vein occlusion and portal hypertension.

BACKGROUND: Our aim was to evaluate the perioperative morbidity and survival of a selected group of patients with locally advanced pancreatic ductal adenocarcinoma (PDAC) and malignant obstruction of portal axis inducing portal hypertension (PH) who underwent a curative intent pancreatic resection, after neoadjuvant chemotherapy, adopting a new type of temporary intraoperative mesentericoportal shunt (TMPS). METHODS: We analyzed the perioperative data and survival outcome of 15 patients with locally advanced PDAC and PH who underwent pancreatoduodenectomy combined with vascular resections between October 2008 and October 2012 using this TMPS. RESULTS: There was no perioperative mortality. Postoperative morbidity occurred in 7 patients without any postoperative liver failure. All patients underwent mesentericoportal venous resection, 11 of whom had a concomitant arterial resection. The mean ± SD follow-up was 16 ± 10 months (range, 4-40; median 15). Overall survival rates of patients were 78% and 11% at 1 and at 3 years, respectively. Median survival was 17 months. The 1-year disease-free survival was 36%. CONCLUSION: The use of this form of TMPS allowed us to achieve PD or total pancreatectomy in patients with locally advanced PDAC and PH without postoperative mortality but with increased morbidity. The relevance of such an aggressive approach is yet to be determined.

Pancreatic fistula after a pancreaticoduodenectomy for ductal adenocarcinoma and its association with morbidity: a multicentre study of the French Surgical Association.

BACKGROUNDS: A pancreatic fistula (PF) is the most relevant complication after a pancreaticoduodenectomy (PD). This retrospective multicentric study attempts to elucidate the risk factors and complications of a PF in a large cohort of patients undergoing a PD for ductal adenocarcinoma. METHODS: Using a survey tool, clinical data of 1325 patients undergoing a PD for ductal adenocarcinoma at 37 institutions, between January 2004 and December 2009, were collected. Peri-operative risk factors associated with PF and its association with morbidity and mortality were assessed. Morbidity and PF were graded according to the ISGPF (International Study group for pancreatic fistula) definition and the Dindo-Clavien classification. RESULTS: Overall PF, mortality, morbidity and relaparotomy rates were 14.3%, 3.8%, 54.4% and 11.7%, respectively. PF occurred more frequently after a pancreaticojejunostomy (PJ) compared with a pancreaticogastrostomy (PG) (16.8% vs. 10.4%; P = 0.0012). Independent risk factors for PF by multivariate analysis were absence of pre-operative diabetes (P = 0.0014), PJ reconstruction (P = 0.0035), soft pancreatic parenchyma (P < 0.0001) and low-volume centre (P = 0.0286). Clinically relevant PF (grade B and C) and severe complications (Dindo-Clavien grade IIIB, IV, V) were significantly more frequent after PJ than PG (71.6% vs. 28.3%; P = 0.030 and 24.8% vs. 19.1%; P = 0.015, respectively). Overall mortality and relaparotomy rates were similar after PG and PJ. CONCLUSIONS: A soft pancreatic parenchyma, the absence of pre-operative diabetes, PJ and low-volume centre are independent risk factors for PF after PD for ductal adenocarcinoma. A significantly higher incidence and clinical severity of PF are associated with PJ.