Pancreatic Surgery in Children: Complex, Safe, and Effective.

OBJECTIVE: The aim of this study was to assess indications for and report outcomes of pancreatic surgery in pediatric patients. BACKGROUND: Indications for pancreatic surgery in children are rare and data on surgical outcomes after pediatric pancreatic surgery are scarce. METHODS: All children who underwent pancreatic surgery at a tertiary hospital specializing in pancreatic surgery between 2003 and 2022 were identified from a prospectively maintained database. Indications, surgical procedures, and perioperative as well as long-term outcomes were analyzed. RESULTS: In total, 73 children with a mean age of 12.8 years (range: 4 mo to 18 y) underwent pancreatic surgery during the observation period. Indications included chronic pancreatitis (n=35), pancreatic tumors (n=27), and pancreatic trauma (n=11). Distal pancreatectomy was the most frequently performed procedure (n=23), followed by pancreatoduodenectomy (n=19), duodenum-preserving pancreatic head resection (n=10), segmental pancreatic resection (n=7), total pancreatectomy (n=3), and others (n=11). Postoperative morbidity occurred in 25 patients (34.2%), including 7 cases (9.6%) with major complications (Clavien-Dindo≥III). There was no postoperative (90-d) mortality. The 5-year overall survival was 90.5%. The 5-year event-free survival of patients with chronic pancreatitis was 85.7%, and 69.0% for patients with pancreatic tumors. CONCLUSION: This is the largest single-center study on pediatric pancreatic surgery in a Western population. Pediatric pancreatic surgery can be performed safely. Centralization in pancreatic centers with high expertise in surgery of adult and pediatric patients is important as it both affords the benefits of pancreatic surgery experience and ensures that surgical management is adapted to the specific needs of children.

Who Do We Fail to Rescue after Pancreatoduodenectomy? Outcomes Among >4000 Procedures Expose Windows of Opportunity.

OBJECTIVE: Our investigation on in-hospital mortality after 4474 pancreatoduodenectomies aimed to identify time-dependent risks as well as windows of opportunity to rescue patients from complications. BACKGROUND: Pancreatoduodenectomy is generally considered a safe procedure with a 1-10% perioperative mortality based on complexity and surgical volume. Yet, patients are susceptible for life-threatening complications particularly with extended resections. Recognition of distinct vulnerabilities over time while patients recover is required to permit focused monitoring, sophisticated resource allocation, and greatest surgical safety. METHODS: Patients who deceased in-hospital after pancreatoduodenectomy between 2003-2021 were retrieved from the institutional pancreatectomy registry and analyzed in detail with respect to their postoperative course. RESULTS: Among 4474 pancreatoduodenectomies, 156 patients deceased in-hospital (3.5%). When assessing root causes of mortality, we observed 3 different clusters of complications which were postpancreatectomy-specific (47.4%), visceral vasculature-associated (25.6%), or cardiopulmonary in origin (23.7%). The median times of root cause onset in the 3 categories were postoperative day (POD) 9, POD 4.5 ( P =0.008) and POD 3 ( P <0.001), and medians of in-hospital mortality were POD 31, POD 18 ( P =0.009) and POD 8 ( P <0.001), respectively. Intervals between root cause onset and mortality varied with medians of 23 days, 11 days ( P =0.017), and 1 days ( P <0.001). The 3 categories were similarly distributed between different types of surgical complexity. CONCLUSION: Postpancreatectomy-specific complications prompt almost half of in-hospital mortalities after pancreatoduodenectomy, with rather long intervals for interventions to prevent failure to rescue. In contrast, visceral vasculature-related events and cardiopulmonary complications dominate early in-hospital mortalities with short intervals until mortality, demanding rigorous management of such events or preoperative conditioning. These data externally validate a previous high-volume initiative and highlight distinct windows of opportunity to optimize perioperative safety.

The PAncreatic Surgery Composite Endpoint PACE - Development and Validation of a Clinically Relevant Endpoint Requiring Lower Sample Sizes.

OBJECTIVE: To provide a composite endpoint in pancreatic surgery. SUMMARY BACKGROUND DATA: Single endpoints in prospective and randomized studies have become impractical due to their low frequency and the marginal benefit of new interventions. METHODS: Data from prospective studies were used to develop (n=1273) and validate (n=544) a composite endpoint based on postoperative pancreatic fistula, post-pancreatectomy hemorrhage as well as reoperation and reinterventions. All patients had pancreatectomies of different extents. The association of the developed PAncreatic surgery Composite Endpoint (PACE) with prolonged length of hospital stay (LOS) >75th percentile and mortality was assessed. A single-institution database was used for external validation (n = 2666). Sample size calculations were made for single outcomes and the composite endpoint. RESULTS: In the internal validation cohort, the PACE demonstrated an AUC of 78.0%, a sensitivity of 90.4% and a specificity of 67.6% in predicting a prolonged LOS. In the external cohort, the AUC was 76.9%, the sensitivity 73.8% and the specificity 80.1%. The 90-day mortality rate was significantly different for patients with a positive versus a negative PACE both in the development and internal validation cohort (5.1% vs 0.9%; P< 0.001), as well as in the external validation cohort (8.5% vs 1.2%, P< 0.001). The PACE enabled sample size reductions of up to 80.5% compared to single outcomes. CONCLUSION: The PACE performed well in predicting prolonged hospital stays and can be used as a standardized and clinically relevant endpoint for future prospective trials enabling lower sample sizes and therefore improved feasibility compared to single outcome parameters.

Preoperative Surgical or Endoscopic Bile Duct Drainage in Pancreatic Cancer.

BACKGROUND: Patients with pancreatic cancer and obstructive jaundice routinely undergo endoscopic stent placement (ES). It is well known that ES causes bacterial contamination and infectious complications after pancreatic resection. OBJECTIVE: To compare short-term outcomes and survival in patients undergoing pancreatic head resection after preoperative ES vs preoperative surgical drainage (SD) via T-tube insertion. METHODS: Patients with obstructive jaundice who underwent SD or ES from 2016 to 2022 were identified from a prospective database. Outcome analyses included microbiological bile contamination, overall morbidity and assessment of the overall complication burden using the Comprehensive Complication Index (CCI). Overall survival was investigated by Kaplan‒Meier analysis. RESULTS: A total of 55 patients with SD were identified and matched with 110 ES patients. After the primary intervention, ES patients experienced more complications (ES: 17.3% vs. SD: 3.6%; P=0.013). The overall complication burden after pancreatic resection was higher in ES patients than in SD patients (CCI: 27.2 vs. 19.9; P=0.022). Additionally, bacterial contamination of the bile was more frequent in ES patients compared to SD individuals (94.3% vs. 7.1%; P<0.001) with similar bacteria in 83.3% of postoperative abdominal infections in ES patients. While overall survival did not differ between the two groups, patients with postinterventional complications after ES had an impaired survival compared to those without complications (11.3 mo vs. 20.4 mo; P=0.03). CONCLUSION: SD for obstructive jaundice in resectable pancreatic cancer is associated with a lower overall complication burden. Additionally, patients with complications after ES experience worse overall survival. These findings indicate to rethink our standards of treatment of obstructive jaundice in patients with pancreatic cancer.

Complexity and Experience Grading to Guide Patient Selection for Minimally-invasive Pancreatoduodenectomy: An ISGPS Consensus.

OBJECTIVE: The ISGPS aims to develop a universally accepted complexity and experience grading system to guide the safe implementation of robotic and laparoscopic minimally-invasive pancreatoduodenectomy (MIPD). BACKGROUND: Despite the perceived advantages of MIPD, its global adoption has been slow due to the inherent complexity of the procedure and challenges to acquiring surgical experience. Its wider adoption must be undertaken with an emphasis towards appropriate patient selection according to adequate surgeon and center experience. METHODS: The ISGPS developed a complexity and experience grading system to guide patient selection for MIPD based on an evidence-based review and a series of discussions. RESULTS: The ISGPS complexity and experience grading system for MIPD is subclassified into patient-related risk factors and provider experience-related variables. The patient-related risk factors include anatomical (main pancreatic and common bile duct diameters), tumor-specific (vascular contact), and conditional (obesity and previous complicated upper abdominal surgery/disease) factors, all incorporated in an A-B-C classification, graded as no, a single, and multiple risk factors. The surgeon and center experience-related variables include surgeon total MIPD experience (cut-offs 40 and 80) and center annual MIPD volume (cut-offs 10 and 30), all also incorporated in an A-B-C classification. CONCLUSION: This ISGPS complexity and experience grading system for robotic and laparoscopic MIPD may enable surgeons to optimally select patients after duly considering specific risk factors known to influence the complexity of the procedure. This grading system will likely allow for a thoughtful and stepwise implementation of MIPD and facilitate a fair comparison of outcome between centers and countries.

Risk Factors for Rectal Stump Leakage After Discontinuity Resection: Stump Length Matters Most.

BACKGROUND: Discontinuity resection is commonly conducted to avoid anastomotic leakage in high-risk patients but potentially results in rectal stump leakage. Although risk factors for anastomotic leakage have been widely studied, data on rectal stump leakage rates and underlying risk factors are scarce. OBJECTIVE: To determine rectal stump leakage rates following Hartmann's procedure and to identify patient-and surgery-associated risk factors. DESIGN: A retrospective study with univariate and multivariate analyses was performed to identify risk factors of rectal stump leakage. A subgroup analysis of scheduled operations was performed. SETTINGS: The study was conducted at Heidelberg University Hospital, Germany. PATIENTS: Patients were included who underwent discontinuity resection with rectal stump formation between 2010 and 2020. MAIN OUTCOME MEASURES: The main outcome measures included rectal stump leakage rates, 30-day mortality, length of hospitalization, and necessity for further invasive treatment. RESULTS: Rectal stump leakage occurred in 11.78% of patients. Rectal stump leakage rates varied considerably depending on the surgical procedure performed and were highest following subtotal pelvic exenteration (34%). Diagnosis of rectal stump leakage peaked on postoperative day 7. A short rectal stump ( p = 0.001), previous pelvic radiotherapy ( p = 0.04), chemotherapy ( p = 0.004), and previous laparotomy ( p = 0.03) were independent risk factors for rectal stump leakage in the entire patient collective. In patients undergoing scheduled surgery, a short rectal stump was the only independent risk factor ( p = 0.003). Rectal stump leakage was not associated with increased 30-day mortality but prolonged length of hospitalization and frequently necessitated further invasive treatment. LIMITATIONS: Study results are limited by the retrospective design, a high number of emergency operations, and the mere inclusion of symptomatic leakages. CONCLUSIONS: Rectal stump leakage is a relevant complication after discontinuity resection. Risk factors should be considered during surgical decision-making when both discontinuity resection and abdominoperineal resection are feasible. See Video Abstract. FACTORES DE RIESGO PARA LA FUGA DEL MUN RECTAL DESPUS DE UNA RESECCIN POR DISCONTINUIDAD LA LONGITUD DEL MUN ES LO MS IMPORTANTE: ANTECEDENTES:La resección de discontinuidad se realiza comúnmente para evitar la fuga anastomótica en pacientes de alto riesgo, pero potencialmente da como resultado una fuga del muñón rectal. Si bien los factores de riesgo de fuga anastomótica se han estudiado ampliamente, los datos sobre las tasas de fuga del muñón rectal y los factores de riesgo subyacentes son escasos.OBJETIVO:Determinar las tasas de fuga del muñón rectal después del procedimiento de Hartmann e identificar los factores de riesgo asociados con el paciente y la cirugía.DISEÑO:Se realizó un estudio retrospectivo con análisis univariado y multivariado para identificar los factores de riesgo de fuga del muñón rectal. Se llevó a cabo un análisis de subgrupos de las operaciones programadas.AJUSTES:El estudio se realizó en el Hospital Universitario de Heidelberg, Alemania.PACIENTES:Se incluyeron pacientes que se sometieron a resección de discontinuidad con formación de muñón rectal entre 2010 y 2020.MEDIDAS DE RESULTADO PRINCIPALES:Las principales medidas de resultado incluyeron las tasas de fuga del muñón rectal, la mortalidad a los 30 días, la duración de la hospitalización y la necesidad de un tratamiento invasivo adicional.RESULTADOS:La fuga del muñón rectal ocurrió en el 11,78% de los pacientes. Las tasas de fuga del muñón rectal variaron considerablemente según el procedimiento quirúrgico realizado y fueron más altas después de la exenteración pélvica subtotal (34%). El diagnóstico de fuga del muñón rectal alcanzó su punto máximo en el día 7 del postoperatorio. Un muñón rectal corto (p = 0,001), radioterapia pélvica previa (p = 0,04), quimioterapia (p = 0,004) y laparotomía previa (p = 0,03) fueron factores de riesgo independientes de fuga rectal. Fuga del muñón en todo el colectivo de pacientes. En los pacientes sometidos a cirugía programada, el muñón rectal corto fue el único factor de riesgo independiente (p = 0,003). La fuga del muñón rectal no se asoció con un aumento de la mortalidad a los 30 días, pero con una duración prolongada de la hospitalización y con frecuencia requirió un tratamiento invasivo adicional.LIMITACIONES:Los resultados del estudio están limitados por el diseño retrospectivo, un alto número de operaciones de emergencia y la mera inclusión de fugas sintomáticas.CONCLUSIONES:La fuga del muñón rectal es una complicación relevante tras la resección por discontinuidad. Se deben considerar los factores de riesgo durante la toma de decisiones quirúrgicas cuando son factibles tanto la resección por discontinuidad como la resección abdominoperineal. (Traducción-Yesenia Rojas-Khalil ).

Outcomes in intraductal papillary mucinous neoplasm-derived pancreatic cancer differ from PanIN-derived pancreatic cancer.

BACKGROUND AND AIM: Intraductal papillary mucinous neoplasm (IPMN)-derived pancreatic ductal adenocarcinoma (PDAC) management is generally extrapolated from pancreatic intraepithelial neoplasia (PanIN)-derived PDAC guidelines. However, these are biologically divergent, and heterogeneity further exists between tubular and colloid subtypes. METHODS: Consecutive upfront surgery patients with PanIN-derived and IPMN-derived PDAC were retrospectively identified from international centers (2000-2019). One-to-one propensity score matching for clinicopathologic factors generated three cohorts: IPMN-derived versus PanIN-derived PDAC, tubular IPMN-derived versus PanIN-derived PDAC, and tubular versus colloid IPMN-derived PDAC. Overall survival (OS) was compared using Kaplan-Meier and log-rank tests. Multivariable Cox regression determined corresponding hazard ratios (HR) and 95% confidence intervals (95% CI). RESULTS: The median OS (mOS) in 2350 PanIN-derived and 700 IPMN-derived PDAC patients was 23.0 and 43.1 months (P < 0.001), respectively. PanIN-derived PDAC had worse T-stage, CA19-9, grade, and nodal status. Tubular subtype had worse T-stage, CA19-9, grade, nodal status, and R1 margins, with a mOS of 33.7 versus 94.1 months (P < 0.001) in colloid. Matched (n = 495), PanIN-derived and IPMN-derived PDAC had mOSs of 30.6 and 42.8 months (P < 0.001), respectively. In matched (n = 341) PanIN-derived and tubular IPMN-derived PDAC, mOS remained poorer (27.7 vs 37.4, P < 0.001). Matched tubular and colloid cancers (n = 112) had similar OS (P = 0.55). On multivariable Cox regression, PanIN-derived PDAC was associated with worse OS than IPMN-derived (HR: 1.66, 95% CI: 1.44-1.90) and tubular IPMN-derived (HR: 1.53, 95% CI: 1.32-1.77) PDAC. Colloid and tubular subtype was not associated with OS (P = 0.16). CONCLUSIONS: PanIN-derived PDAC has worse survival than IPMN-derived PDAC supporting distinct outcomes. Although more indolent, colloid IPMN-derived PDAC has similar survival to tubular after risk adjustment.

Tranexamic Acid Administration During Liver Transplantation Is Not Associated With Lower Blood Loss or With Reduced Utilization of Red Blood Cell Transfusion.

BACKGROUND: Current clinical guidelines recommend antifibrinolytic treatment for liver transplantation to reduce blood loss and transfusion utilization. However, the clinical relevance of fibrinolysis during liver transplantation is questionable, a benefit of tranexamic acid (TXA) in this context is not supported by sufficient evidence, and adverse effects are also conceivable. Therefore, we tested the hypothesis that use of TXA is associated with reduced blood loss. METHODS: We performed a retrospective cohort study on patients who underwent liver transplantation between 2004 and 2017 at Heidelberg University Hospital, Heidelberg, Germany. Univariable and multivariable linear regression analyses were used to determine the association between TXA administration and the primary end point intraoperative blood loss and the secondary end point intra- and postoperative red blood cell (RBC) transfusions. For further secondary outcome analyses, the time to the first occurrence of a composite end point of hepatic artery thrombosis, portal vein thrombosis, and thrombosis of the inferior vena cava were analyzed using a univariable and multivariable Cox proportional hazards model. RESULTS: Data from 779 transplantations were included in the final analysis. The median intraoperative blood loss was 3000 mL (1600-5500 mL). Intraoperative TXA administration occurred in 262 patients (33.6%) with an average dose of 1.4 ± 0.7 g and was not associated with intraoperative blood loss (regression coefficient B, -0.020 [-0.051 to 0.012], P = .226) or any of the secondary end points (intraoperative RBC transfusion; regression coefficient B, 0.023 [-0.006 to 0.053], P = .116), postoperative RBC transfusion (regression coefficient B, 0.007 [-0.032 to 0.046], P = .717), and occurrence of thrombosis (hazard ratio [HR], 1.110 [0.903-1.365], P = .321). CONCLUSIONS: Our data do not support the use of TXA during liver transplantation. Physicians should exercise caution and consider individual factors when deciding whether or not to administer TXA.

Thoracic Epidural Analgesia Is Not Associated With Improved Survival After Pancreatic Surgery: Long-Term Follow-Up of the Randomized Controlled PAKMAN Trial.

BACKGROUND: Perioperative thoracic epidural analgesia (EDA) and patient-controlled intravenous analgesia (PCIA) are common forms of analgesia after pancreatic surgery. Current guidelines recommend EDA over PCIA, and evidence suggests that EDA may improve long-term survival after surgery, especially in cancer patients. The aim of this study was to determine whether perioperative EDA is associated with an improved patient prognosis compared to PCIA in pancreatic surgery. METHODS: The PAKMAN trial was an adaptive, pragmatic, international, multicenter, randomized controlled superiority trial conducted from June 2015 to October 2017. Three to five years after index surgery a long-term follow-up was performed from October 2020 to April 2021. RESULTS: For long-term follow-up of survival, 109 patients with EDA were compared to 111 patients with PCIA after partial pancreatoduodenectomy (PD). Long-term follow-up of quality of life (QoL) and pain assessment was available for 40 patients with EDA and 45 patients with PCIA (questionnaire response rate: 94%). Survival analysis revealed that EDA, when compared to PCIA, was not associated with improved overall survival (OS, HR, 1.176, 95% HR-CI, 0.809-1.710, P = .397, n = 220). Likewise, recurrence-free survival did not differ between groups (HR, 1.116, 95% HR-CI, 0.817-1.664, P = .397, n = 220). OS subgroup analysis including only patients with malignancies showed no significant difference between EDA and PCIA (HR, 1.369, 95% HR-CI, 0.932-2.011, P = .109, n = 179). Similar long-term effects on QoL and pain severity were observed in both groups (EDA: n = 40, PCIA: n = 45). CONCLUSIONS: Results from this long-term follow-up of the PAKMAN randomized controlled trial do not support favoring EDA over PCIA in pancreatic surgery. Until further evidence is available, EDA and PCIA should be considered similar regarding long-term survival.

Portal vein resection in pancreatic neuroendocrine neoplasms.

BACKGROUND: Surgery offers the only cure for borderline resectable or locally advanced pancreatic neuroendocrine neoplasms. Data on incidence, perioperative and long-term outcomes of portal vein resection for pancreatic neuroendocrine neoplasms are scarce. This study aimed to analyze the outcome and prognostic factors of portal vein resection in surgery for pancreatic neuroendocrine neoplasms. METHODS: Consecutive patients were analyzed. Portal vein resection was classified according to the International Study Group of Pancreatic Surgery. Clinicopathologic features and overall and disease-free survival were assessed and compared with standard resection in a matched-pair analysis. RESULTS: A total of 54 of 666 (8%) resected pancreatic neuroendocrine neoplasms patients underwent portal vein resection, including 7 (13%) tangential resections with venorrhaphy (type 1), 2 (4%) patch reconstructions (type 2), 35 (65%) end-to-end anastomoses (type 3), and 10 (19%) graft interpositions (type 4); 52% of those underwent pancreatoduodenectomy, 22% distal pancreatectomy, and 26% total pancreatectomy. Postoperative portal vein thrombosis occurred in 19%. Postoperative pancreatic fistula grades B and C (9% vs 16%; P = .357), complications Clavien-Dindo grade ≥IIIb (28% vs 13%; P = .071), and 90-day mortality rate (2% each) were not significantly different compared with 108 matched patients. The 5-year overall survival was 45% (standard resection: 68%; P = .432), and the 5-year disease-free survival was 25% (standard resection: 34%; P = .716). Radical resection was associated with 5-year overall survival of 51% and 5-year disease-specific survival of 75%. CONCLUSION: This is the largest single-center analysis evaluating perioperative and long-term outcomes of portal vein resection for pancreatic neuroendocrine neoplasms. The postoperative complication rate after portal vein resection is comparable with standard resection. The 90-day mortality is low. Radical resection leads to excellent 5-year oncological survival.

Low intraoperative end-tidal carbon dioxide levels are associated with improved recurrence-free survival after elective colorectal cancer surgery.

STUDY OBJECTIVE: Higher levels of carbon dioxide (CO2) increase the invasive abilities of colon cancer cells in vitro. Studies assessing target values for end-tidal CO2 concentrations (EtCO2) to improve surgical outcome after colorectal cancer surgery are lacking. Therefore, we evaluated whether intraoperative EtCO2 was associated with differences in recurrence-free survival after elective colorectal cancer (CRC) surgery. DESIGN: Single center, retrospective analysis. SETTING: Anesthesia records, surgical databases and hospital information system of a tertiary university hospital. PATIENTS: We analyzed 528 patients undergoing elective resection of colorectal cancer at Heidelberg University Hospital between 2009 and 2018. INTERVENTIONS: None. MEASUREMENTS: Intraoperative mean EtCO2 values were calculated. The study cohort was equally stratified into low-and high-EtCO2 groups. The primary endpoint measure was recurrence-free survival until last known follow-up. Groups were compared using Kaplan-Meier analysis. Cox-regression analysis was used to control for covariates. Sepsis, reoperations, surgical site infections and cardiovascular events during hospital stay, and overall survival were secondary outcomes. MAIN RESULTS: Mean EtCO2 was 33.8 mmHg ±1.2 in the low- EtCO2 group vs. 37.3 mmHg ±1.6 in the high-EtCO2 group. Median follow-up was 3.8 (Q1-Q3, 2.5-5.1) years. Recurrence-free survival was higher in the low-EtCO2 group (log-rank-test: p = .024). After correction for confounding factors, lower EtCO2 was associated with increased recurrence-free survival (HR = 1.138, 95%-CI:1.015-1.276, p = .027); the hazard for the primary outcome decreased by 12.1% per 1 mmHg decrease in mean EtCO2. 1-year and 5-year survival was also higher in the low-EtCO2 group. We did not find differences in the other secondary endpoints. CONCLUSIONS: Lower intraoperative EtCO2 target values in CRC surgery might benefit oncological outcome and should be evaluated in confirmative studies.

Robotic versus open partial pancreatoduodenectomy (EUROPA): a randomised controlled stage 2b trial.

Background: Open partial pancreatoduodenectomy (OPD) represents the current gold standard of surgical treatment of a wide range of diseases of the pancreatic head but is associated with morbidity in around 40% of cases. Robotic partial pancreatoduodenectomy (RPD) is being used increasingly, yet, no randomised controlled trials (RCTs) of RPD versus OPD have been published, leaving a low level of evidence to support this practice. Methods: This investigator-initiated, exploratory RCT with two parallel study arms was conducted at a high-volume pancreatic centre in line with IDEAL recommendations (stage 2b). Patients scheduled for elective partial pancreatoduodenectomy (PD) for any indication were randomised (1:1) to RPD or OPD with a centralised web-based tool. The primary endpoint was postoperative cumulative morbidity within 90 days, assessed via the Comprehensive Complication Index (CCI). Biometricians were blinded to the intervention, but patients and surgeons were not. The trial was registered prospectively (DRKS00020407). Findings: Between June 3, 2020 and February 14, 2022, 81 patients were randomly assigned to RPD (n = 41) or OPD (n = 40), of whom 62 patients (RPD: n = 29, OPD: n = 33) were analysed in the modified intention to treat analysis. Four patients in the OPD group were randomised, but did not undergo surgery in our department and one patient was excluded in the RPD group due to other reason. Nine patients in the RPD group and 3 patients in the OPD were excluded from the primary analysis because they did not undergo PD, but rather underwent other types of surgery. The CCI after 90 days was comparable between groups (RPD: 34.02 ± 23.48 versus OPD: 36.45 ± 27.65, difference in means [95% CI]: -2.42 [-15.55; 10.71], p = 0.713). The RPD group had a higher incidence of grade B/C pancreas-specific complications compared to the OPD group (17 (58.6%) versus 11 (33.3%); difference in rates [95% CI]: 25.3% [1.2%; 49.4%], p = 0.046). The only complication that occurred significantly more often in the RPD than in the OPD group was clinically relevant delayed gastric emptying. Procedure-related and overall hospital costs were significantly higher and duration of surgery was longer in the RPD group. Blood loss did not differ significantly between groups. The intraoperative conversion rate of RPD was 23%. Overall 90-day mortality was 4.8% without significant differences between RPD and OPD. Interpretation: In the setting of a very high-volume centre, both RPD and OPD can be considered safe techniques. Further confirmatory multicentre RCTs are warranted to uncover potential advantages of RPD in terms of perioperative and long-term outcomes. Funding: Federal Ministry of Education and Research (BMBF: 01KG2010).

Primary Tumor Resection Before Systemic Therapy in Patients With Colon Cancer and Unresectable Metastases: Combined Results of the SYNCHRONOUS and CCRe-IV Trials.

PURPOSE: Chemotherapy is established as primary treatment in patients with stage IV colorectal cancer and unresectable metastases. Data from nonrandomized clinical trials have fueled persistent uncertainty if primary tumor resection (PTR) before chemotherapy prolongs survival. We investigated the prognostic value of PTR in patients with newly diagnosed stage IV colon cancer who were not amenable to curative treatment. PATIENTS AND METHODS: Patients enrolled in the multicenter, randomized SYNCHRONOUS and CCRe-IV trials were included in the analysis. Patients with colon cancer with synchronous unresectable metastases were randomly assigned at 100 sites in Austria, Germany, and Spain to undergo PTR or up-front chemotherapy (No PTR group). The chemotherapy regimen was left at discretion of the local team. Patients with tumor-related symptoms, inability to tolerate surgery and/or systemic chemotherapy, and history of another cancer were excluded. The primary end point was overall survival (OS), and the analyses were performed with intention-to-treat. RESULTS: A total of 393 patients were randomly assigned to undergo PTR (n = 187) or no PTR (n = 206) between November 2011 and March 2017. Chemotherapy was not administered to 6.4% in the No PTR group and 24.1% in the PTR group. The median follow-up time was 36.7 months (95% CI, 36.6 to 37.3). The median OS was 16.7 months (95% CI, 13.2 to 19.2) in the PTR group and 18.6 months (95% CI, 16.2 to 22.3) in the No PTR group (P = .191). Comparable OS between the study groups was further confirmed on multivariate analysis (hazard ratio, 0.944 [95% CI, 0.738 to 1.209], P = .65) and across all subgroups. Patients with serious adverse events were more common in the No PTR group (10.2% v 18.0%; P = .027). CONCLUSION: Among patients with colon cancer and synchronous unresectable metastases, PTR before systemic chemotherapy was not associated with prolonged OS.

An international Multi-Institutional validation of T1 Sub-staging of intraductal papillary mucinous neoplasm-derived pancreatic cancer.

BACKGROUND: Intraductal papillary mucinous neoplasm (IPMN)-derived pancreatic ductal adenocarcinoma (PDAC) is resected at smaller sizes compared to its biologically distinct counterpart, pancreatic intraepithelial neoplasia (PanIN)-derived PDAC. Thus, experts proposed T1 sub-staging for IPMN-derived PDAC. However, this has never been validated. METHODS: Consecutive upfront surgery patients with IPMN-derived PDAC from five international high-volume centers were classified by the proposed T1 sub-staging classification (T1a ≤ 0.5, T1b > 0.5 and ≤1.0, and T1c >1.0 and ≤2.0 cm) using the invasive component size. Kaplan-Meier and log-rank tests were utilized to compare overall survival (OS). A multivariable Cox-regression was used to determine hazard ratios (HR) with confidence intervals (95%CI). RESULTS: Among 747 patients, 69 (9.2%), 50 (6.7%), 99 (13.0%), and 531 patients (71.1%), comprised the T1a, T1b, T1c, and T2-4 subgroups, respectively. Increasing T-stage was associated with elevated CA19-9, poorer grade, nodal positivity, R1-margin, and tubular subtype. Median OS for T1a, T1b, T1c, and T2-4 were 159.0 (95%CI:126.0-NR), 128.8 (98.3-NR), 77.6 (48.3-108.2), and 31.4 (27.5-37.7) months, respectively (p < .001). OS decreased with increasing T-stage for all pairwise comparisons (all p < .05). After risk-adjustment, age > 65, elevated CA19-9, T1b [HR : 2.55 (1.22-5.32)], T1c [HR : 3.04 (1.60-5.76)], and T2-4 [HR : 3.41 (1.89-6.17)] compared to T1a, nodal positivity, R1-margin, and no adjuvant chemotherapy were associated with worse OS. Disease recurrence was more common in T2-4 tumors (56.4%) compared to T1a (18.2%), T1b (23.9%), and T1c (36.1%, p < .001). CONCLUSION: T1 sub-staging of IPMN-derived PDAC is valid and has significant prognostic value. Advancing T1 sub-stage is associated with worse histopathology, survival, and recurrence. T1 sub-staging is recommended for future guidelines.

Deep Learning Histology for Prediction of Lymph Node Metastases and Tumor Regression after Neoadjuvant FLOT Therapy of Gastroesophageal Adenocarcinoma.

BACKGROUND: The aim of this study was to establish a deep learning prediction model for neoadjuvant FLOT chemotherapy response. The neural network utilized clinical data and visual information from whole-slide images (WSIs) of therapy-naïve gastroesophageal cancer biopsies. METHODS: This study included 78 patients from the University Hospital of Cologne and 59 patients from the University Hospital of Heidelberg used as external validation. RESULTS: After surgical resection, 33 patients from Cologne (42.3%) were ypN0 and 45 patients (57.7%) were ypN+, while 23 patients from Heidelberg (39.0%) were ypN0 and 36 patients (61.0%) were ypN+ (p = 0.695). The neural network had an accuracy of 92.1% to predict lymph node metastasis and the area under the curve (AUC) was 0.726. A total of 43 patients from Cologne (55.1%) had less than 50% residual vital tumor (RVT) compared to 34 patients from Heidelberg (57.6%, p = 0.955). The model was able to predict tumor regression with an error of ±14.1% and an AUC of 0.648. CONCLUSIONS: This study demonstrates that visual features extracted by deep learning from therapy-naïve biopsies of gastroesophageal adenocarcinomas correlate with positive lymph nodes and tumor regression. The results will be confirmed in prospective studies to achieve early allocation of patients to the most promising treatment.

Impact of circumferential resection margin on survival in ampullary cancer: retrospective analysis.

BACKGROUND: Ampullary carcinoma is a clinically variable entity. This study aimed to evaluate prognostic factors for the outcome of resected ampullary carcinoma patients with particular intent to analyse the influence of surgical radicality. METHODS: Patients undergoing resection between 2002 and 2017 were analysed. Clinicopathological parameters, perioperative outcome and survival were examined. Risk factor analysis for postresection survival was performed. Resection margin status was evaluated according to the revised classification for pancreatic adenocarcinoma. RESULTS: A total of 234 patients were identified, 97.9 per cent (n = 229) underwent formal resection, while 2.1 per cent (n = 5) underwent ampullary resection. Histological subtypes were 46.6 per cent (n = 109) pancreatobiliary, 34.2 per cent (n = 80) intestinal, 11.5 per cent (n = 27) mixed, and 7.7 per cent (n = 18) undetermined. In the pancreatobiliary group, tumours were more advanced with more vascular resections, pT4 stage, G3 differentiation and pN+ status. Five-year overall survival was significantly different for pancreatobiliary compared to intestinal (51.7 per cent versus 72.8 per cent, P = 0.0087). In univariable analysis, age, pT4 stage, pN+, pancreatobiliary subtype and positive resection margin were significantly associated with worse overall survival. Long-term outcome was significantly better after true R0 resection (circumferential resection margin-, tumour clearance >1 mm) compared with circumferential resection margin+ (<1 mm) and R1 resections (5-year overall survival: 69.6 per cent, median overall survival 191 months versus 42.4 per cent and 53 months; P = 0.0017). CONCLUSION: Postresection survival of ampullary carcinoma patients is determined by histological subtype and surgical radicality. Intestinal differentiation is associated with less advanced tumour stages and better differentiation, which is reflected in a significantly better overall survival compared to pancreatobiliary differentiation. Despite this, true R0-resection is a prognostic key determinant in both entities, achieving 5-year survival in two-thirds of patients.