Cost-effectiveness of robotic vs laparoscopic distal pancreatectomy. Results from the national prospective trial ROBOCOSTES.

INTRODUCTION: Although several studies report that the robotic approach is more costly than laparoscopy, the cost-effectiveness of robotic distal pancreatectomy (RDP) over laparoscopic distal pancreatectomy (LDP) is still an issue. This study evaluates the cost-effectiveness of the RDP and LDP approaches across several Spanish centres. METHODS: This study is an observational, multicenter, national prospective study (ROBOCOSTES). For one year from 2022, all consecutive patients undergoing minimally invasive distal pancreatectomy were included, and clinical, QALY, and cost data were prospectively collected. The primary aim was to analyze the cost-effectiveness between RDP and LDP. RESULTS: During the study period, 80 procedures from 14 Spanish centres were analyzed. LDP had a shorter operative time than the RDP approach (192.2 min vs 241.3 min, p = 0.004). RDP showed a lower conversion rate (19.5% vs 2.5%, p = 0.006) and a lower splenectomy rate (60% vs 26.5%, p = 0.004). A statistically significant difference was reported for the Comprehensive Complication Index between the two study groups, favouring the robotic approach (12.7 vs 6.1, p = 0.022). RDP was associated with increased operative costs of 1600 euros (p < 0.031), while overall cost expenses resulted in being 1070.92 Euros higher than the LDP but without a statistically significant difference (p = 0.064). The mean QALYs at 90 days after surgery for RDP (0.9534) were higher than those of LDP (0.8882) (p = 0.030). At a willingness-to-pay threshold of 20,000 and 30,000 euros, there was a 62.64% and 71.30% probability that RDP was more cost-effective than LDP, respectively. CONCLUSIONS: The RDP procedure in the Spanish healthcare system appears more cost-effective than the LDP.

Pancreaticoduodenectomy in a patient with severe portal hypertension: the role of preoperative TIPS.

The complication rate for Pancreaticoduodenectomy (PD) is 40-50% in most published series and mortality can raise up to 4-5% even in high-volume centers. Severe portal hypertension secondary to liver disease is associated to high perioperative mortality and therefore is considered a contraindication for PD. No standardized management exists for surgically resectable patients with periampullary cancer and severe portal hypertension. The aim of this case study is to analyse the treatment alternatives in patients with periampullary cancer and severe portal hypertension and focus into the surgical treatment of these patients. We present the case of a 67 year-old patient case with a resectable ampullary cancer and portal hypertension managed with Preoperative Transjugular Intrahepatic Portosystemic Shunt (TIPS) to allow a PD. We present a literature review on the use of preoperative TIPS in patients who are candidates to PD. Neoadjuvant TIPS can be safely used in selected patients with severe portal hypertension who need a PD.

Five-year recurrence/survival after pancreatoduodenectomy for pancreatic adenocarcinoma: does pre-existing diabetes matter? Results from the Recurrence After Whipple's (RAW) study.

BACKGROUND: Diabetes mellitus (DM) has a complex relationship with pancreatic cancer. This study examines the impact of preoperative DM, both recent-onset and pre-existing, on long-term outcomes following pancreatoduodenectomy (PD) for pancreatic ductal adenocarcinoma (PDAC). METHODS: Data were extracted from the Recurrence After Whipple's (RAW) study, a multi-centre cohort of PD for pancreatic head malignancy (2012-2015). Recurrence and five-year survival rates of patients with DM were compared to those without, and subgroup analysis performed to compare patients with recent-onset DM (less than one year) to patients with established DM. RESULTS: Out of 758 patients included, 187 (24.7%) had DM, of whom, 47 of the 187 (25.1%) had recent-onset DM. There was no difference in the rate of postoperative pancreatic fistula (DM: 5.9% vs no DM 9.8%; p = 0.11), five-year survival (DM: 24.1% vs no DM: 22.9%; p = 0.77) or five-year recurrence (DM: 71.7% vs no DM: 67.4%; p = 0.32). There was also no difference between patients with recent-onset DM and patients with established DM in postoperative outcomes, recurrence, or survival. CONCLUSION: We found no difference in five-year recurrence and survival between diabetic patients and those without diabetes. Patients with pre-existing DM should be evaluated for PD on a comparable basis to non-diabetic patients.

Conversion to open surgery in obese patients undergoing minimally invasive distal pancreatectomy: results from a multicenter analysis.

BACKGROUND: Although minimally invasive distal pancreatectomy (MIDP) is considered a standard approach it still presents a non-negligible rate of conversion to open that is mainly related to some difficulty factors, as obesity. The aim of this study is to analyze the preoperative factors associated with conversion in obese patients with MIDP. METHODS: In this multicenter study, all obese patients who underwent MIDP at 18 international expert centers were included. The preoperative factors associated with conversion to open surgery were analyzed. RESULTS: Out of 436 patients, 91 (20.9%) underwent conversion to open, presenting higher blood loss, longer operative time and similar rate of major complications. Twenty (22%) patients received emergent conversion. At univariate analysis, the type of approach, radiological invasion of adjacent organs, preoperative enlarged lymphnodes and ASA ≥ III were significantly associated with conversion to open. At multivariate analysis, robotic approach showed a significantly lower conversion rate (14.6 % vs 27.3%, OR = 2.380, p = 0.001). ASA ≥ III (OR = 2.391, p = 0.002) and preoperative enlarged lymphnodes (OR = 3.836, p = 0.003) were also independently associated with conversion. CONCLUSION: Conversion rate is significantly lower in patients undergoing robotic approach. Radiological enlarged lymphnodes and ASA ≥ III are also associated with conversion to open. Conversion is associated with poorer perioperative outcomes, especially in case of intraoperative hemorrhage.

Robotic versus laparoscopic distal pancreatectomy in obese patients.

BACKGROUND: Although robotic distal pancreatectomy (RDP) has a lower conversion rate to open surgery and causes less blood loss than laparoscopic distal pancreatectomy (LDP), clear evidence on the impact of the surgical approach on morbidity is lacking. Prior studies have shown a higher rate of complications among obese patients undergoing pancreatectomy. The primary aim of this study is to compare short-term outcomes of RDP vs. LDP in patients with a BMI ≥ 30. METHODS: In this multicenter study, all obese patients who underwent RDP or LDP for any indication between 2012 and 2022 at 18 international expert centers were included. The baseline characteristics underwent inverse probability treatment weighting to minimize allocation bias. RESULTS: Of 446 patients, 219 (50.2%) patients underwent RDP. The median age was 60 years, the median BMI was 33 (31-36), and the preoperative diagnosis was ductal adenocarcinoma in 21% of cases. The conversion rate was 19.9%, the overall complication rate was 57.8%, and the 90-day mortality rate was 0.7% (3 patients). RDP was associated with a lower complication rate (OR 0.68, 95% CI 0.52-0.89; p = 0.005), less blood loss (150 vs. 200 ml; p < 0.001), fewer blood transfusion requirements (OR 0.28, 95% CI 0.15-0.50; p < 0.001) and a lower Comprehensive Complications Index (8.7 vs. 8.9, p < 0.001) than LPD. RPD had a lower conversion rate (OR 0.27, 95% CI 0.19-0.39; p < 0.001) and achieved better spleen preservation rate (OR 1.96, 95% CI 1.13-3.39; p = 0.016) than LPD. CONCLUSIONS: In obese patients, RDP is associated with a lower conversion rate, fewer complications and better short-term outcomes than LPD.

Percutaneous transhepatic biliary drainage for biliary obstruction in perihilar cholangiocarcinoma: a 10-year analysis of safety and outcomes using the CCI index.

PURPOSE: To evaluate percutaneous transhepatic biliary drainage (PTBD) safety and efficacy in patients with perihilar cholangiocarcinoma (PCCA). METHODS: This retrospective observational study included patients with PCCA and obstructive cholestasis referred for a PTBD in our institution between 2010 and 2020. Technical and clinical success rates and major complication and mortality rates one month after PTBD were used as main variables. Patients were divided and analyzed into two groups: > 30 and < 30 Comprehensive Complication Index (CCI). We also evaluated post-surgical outcomes in patients undergoing surgery. RESULTS: Out of 223 patients, 57 were included. Technical success rate was 87.7%. Clinical success at 1 week was 83.6%, before surgery 68.2%, 80.0% at 2 weeks and 86.7% at 4 weeks. Mean total bilirubin (TBIL) values were 15.1 mg/dL (baseline), 8.1 mg/dL one week after PTBD), 6.1 mg/dL (2 weeks) and 2.1 mg/dL (4 weeks). Major complication rate was 21.1%. Three patients died (5.3%). Risk factors for major complications after the statistical analysis were: Bismuth classification (p = 0.01), tumor resectability (p = 0.04), PTBD clinical success (p = 0.04), TBIL 2 weeks after PTBD (p = 0.04), a second PTBD (p = 0.01), total PTBDs (p = 0.01) and duration of drainage (p = 0.03). Major postoperative complication rate in patients who underwent surgery was 59.3%, with a median CCI of 26.2. CONCLUSION: PTBD is safe and effective in the management of biliary obstruction caused by PCCA. Bismuth classification, locally advanced tumors, and failure to achieve clinical success in the first PTBD are factors related to major complications. Our sample reported a high major postoperative complication rate, although with an acceptable median CCI.

Intraoperative indocyanine green clearance in liver surgery: a useful tool to promote parenchymal preservation?

A 66-year-old woman was diagnosed with sigmoid carcinoma with bilobar unresectable liver metastases. Primary tumor resection was performed. Neoadjuvant chemotherapy was administered to downstage liver disease. Following FOLFOX/panitumumab (4 cycles), disease progression was observed within the liver. Therefore, chemotherapy regimen was switched to FOLFIRI/cetuximab (12 cycles). At restaging, 7 out of 10 metastases showed complete radiologic response and 3 showed disease progression. Selective internal radiation therapy (SIRT) with Yttrium-90 was performed on these 3 metastases with very good local radiologic response. Surgical resection with curative intent was attempted.  ICG was administered 72 hours before surgery to help localization of liver metastases. Intraoperatively, very poor ICG clearance was demonstrated and therefore parenchyma-sparing resections were performed (segment II, III, IV and VII). At final histology, major pathologic response was observed with steatosis of the liver parenchyma. Postoperative course was uneventful and adjuvant capecitabine was administered (8 cycles). No recurrence was demonstrated at 38 months follow-up. However, eighteen months following surgery, she developed impairment of hepatic function and portal hypertension. Conclusion: Preoperative ICG administration could be helpful to intraoperatively detect patients that can develop late post hepatectomy impairment of hepatic function, especially following SIRT2 (REF2), and promote parenchymal preservation.

Predictors of actual five-year survival and recurrence after pancreatoduodenectomy for ampullary adenocarcinoma: results from an international multicentre retrospective cohort study.

BACKGROUND: Pancreatoduodenectomy (PD) is recommended in fit patients with a resectable ampullary adenocarcinoma (AA). We aimed to identify predictors of five-year recurrence/survival. METHODS: Data were extracted from the Recurrence After Whipple's (RAW) study, a multicentre retrospective study of PD patients with a confirmed head of pancreas or periampullary malignancy (June 1st, 2012-May 31st, 2015). Patients with AA who developed recurrence/died within five-years were compared to those who did not. RESULTS: 394 patients were included and actual five-year survival was 54%. Recurrence affected 45% and the median time-to-recurrence was 14 months. Local only, local and distant, and distant only recurrence affected 34, 41 and 94 patients, respectively (site unknown: 7). Among those with recurrence, the most common sites were the liver (32%), local lymph nodes (14%) and lung/pleura (13%). Following multivariable tests, number of resected nodes, histological T stage > II, lymphatic invasion, perineural invasion (PNI), peripancreatic fat invasion (PPFI) and ≥1 positive resection margin correlated with increased recurrence and reduced survival. Furthermore, ≥1 positive margin, PPFI and PNI were all associated with reduced time-to-recurrence. CONCLUSIONS: This multicentre retrospective study of PD outcomes identified numerous histopathological predictors of AA recurrence. Patients with these high-risk features might benefit from adjuvant therapy.

Multifactorial mitigation strategy to reduce clinically relevant pancreatic fistula in high-risk pancreatojejunostomy following pancreaticoduodenectomy.

INTRODUCTION: Postoperative pancreatic fistula (POPF) is the most dreadful complication of pancreaticoduodenectomy (PD) and previous literature focused on technical modifications of pancreatic remnant reconstruction. We developed a multifactorial mitigation strategy (MS) and the aim of the study is to assess its clinical impact in patients at high-risk of POPF. METHODS: All patients candidate to PD between 2012 and 2018 were considered. Only patients with a high Fistula Risk Score (FRS 7-10) were included. Patients undergoing MS were compared to patients receiving Standard Strategy (SS). Clinical outcomes were compared between the two groups. Multivariate hierarchical logistic regression analyses were performed to detect independent predictors of POPF. RESULTS: Out of 212 patients, 33 were finally included in MS Group and 29 in SS Group. POPF rate was significantly lower in MS Group (12.1% vs 44.8%, p = 0.005). Delayed gastric emptying, postoperative pancreatitis, complications and hospital stay were also significantly lower in MS Group. Hierarchical logistic regression analyses showed that Body Mass Index (OR = 1.196, p = 0.036) and MS (OR = 0.187, p = 0.032) were independently associated with POPF. CONCLUSION: A multifactorial MS can be helpful to reduce POPF rate in patients with high FRS following PD. Personalized approach for vulnerable patients should be investigated in the future.

Clinical impact of preoperative tumour contact with superior mesenteric-portal vein in patients with resectable pancreatic head cancer.

INTRODUCTION: The NCCN classification of resectability in pancreatic head cancer does not consider preoperative radiological tumour ≤ 180° contact with portal vein/superior mesenteric vein (PV/SMV) as a negative prognostic feature. The aim of this study is to evaluate whether this factor is associated with higher rate of incomplete resection and poorer survival. METHODS: All patients considered for pancreatic resection between 2012 and 2017 at two Spanish referral centres were included. Patients with borderline and locally advanced pancreatic ductal adenocarcinoma (PDAC) according to NCCN classification were excluded. Preoperative CT scans were reviewed by dedicated radiologists to identify radiologic tumour contact with PV/SMV. RESULTS: Out of 302, 71 patients were finally included in this study. Twenty-two (31%) patients showed tumour-PV/SMV contact (group 1) and 49 (69%) did not show any contact (group 2). Patients in group 1 showed a statistically significantly higher rate of R1 and R1-direct margins compared with group 2 (95 vs 28% and 77 vs 10%) and lower median survival (24 vs 41 months, p = 0.02). Preoperative contact with PV/SMV, lymph node metastases, R1-direct margin and NO adjuvant chemotherapy were significantly associated with disease-specific survival at multivariate analysis. CONCLUSION: Preoperative radiological tumour contact with PV/SMV in patients with NCCN resectable PDAC is associated with high rate of pathologic positive margins following surgery and poorer survival.

Does the Artery-first Approach Improve the Rate of R0 Resection in Pancreatoduodenectomy?: A Multicenter, Randomized, Controlled Trial.

OBJECTIVE: To compare the rates of R0 resection in pancreatoduodenectomy (PD) for pancreatic and periampullary malignant tumors by means of standard (ST-PD) versus artery-first approach (AFA-PD). BACKGROUND: Standardized histological examination of PD specimens has shown that most pancreatic resections thought to be R0 resections are R1. "Artery-first approach" is a surgical technique characterized by meticulous dissection of arterial planes and clearing of retropancreatic tissue in an attempt to achieve a higher rate of R0. To date, studies comparing AFA-PD versus ST-PD are retrospective cohort or case-control studies. METHODS: A multicenter, randomized, controlled trial was conducted in 10 University Hospitals (NCT02803814, ClinicalTrials.gov). Eligible patients were those who presented with pancreatic head adenocarcinoma and periampullary tumors (ampulloma, distal cholangiocarcinoma, duodenal adenocarcinoma). Assignment to each group (ST-PD or AFA-PD) was randomized by blocks and stratified by centers. The primary end-point was the rate of tumor-free resection margins (R0); secondary end-points were postoperative complications and mortality. RESULTS: One hundred seventy-nine patients were assessed for eligibility and 176 randomized. After exclusions, the final analysis included 75 ST-PD and 78 AFA-PD. R0 resection rates were 77.3% (95% CI: 68.4-87.4) with ST-PD and 67.9% (95% CI: 58.3-79.1) with AFA-PD, P=0.194. There were no significant differences in postoperative complication rates, overall 73.3% versus 67.9%, and perioperative mortality 4% versus 6.4%. CONCLUSIONS: Despite theoretical oncological advantages associated with AFA-PD and evidence coming from low-level studies, this multicenter, randomized, controlled trial has found no difference neither in R0 resection rates nor in postoperative complications in patients undergoing ST-PD versus AFA-PD for pancreatic head adenocarcinoma and other periampullary tumors.

Somatostatin and the "Small-For-Size" Liver.

"Small-for-size" livers arising in the context of liver resection and transplantation are vulnerable to the effects of increased portal flow in the immediate postoperative period. Increased portal flow is an essential stimulus for liver regeneration. If the rise in flow and stimulus for regeneration are excessive; however, liver failure and patient death may result. Somatostatin is an endogenous peptide hormone that may be administered exogenously to not only reduce portal blood flow but also offer direct protection to different cells in the liver. In this review article, we describe key changes that transpire in the liver following a relative size reduction occurring in the context of resection and transplantation and the largely beneficial effects that peri-operative somatostatin therapy may help achieve in this setting.

Prehabilitation in patients undergoing pancreaticoduodenectomy: a randomized controlled trial.

INTRODUCTION: prehabilitation has been proposed as an effective tool to prevent postoperative complications in patients undergoing major abdominal surgery. However, no studies have demonstrated its effectiveness in pancreatic surgical patients. The aim of this study was to assess the impact of prehabilitation on postoperative complications in patients undergoing a pancreaticoduodenectomy (PD). METHODS: this was a randomized controlled trial. Eligible candidates who accepted to participate were randomized to the control (standard care) or intervention (standard care + prehabilitation) group. All patients with pancreatic or periampullary tumors who were candidates for pancreaticoduodenectomy were included. Patients who received neoadjuvant treatment were excluded. Prehabilitation covered three actions: a) nutritional support; b) control of diabetes and exocrine pancreatic insufficiency; and c) physical and respiratory training. The main study outcome was the proportion of patients who suffered postoperative complications. Secondary outcomes included the occurrence of specific complications (pancreatic leak and delayed gastric emptying) and hospital stay. RESULTS: forty patients were included in the analysis. Twenty-two patients were randomized to the control arm and 18, to the intervention group. No statistically significant differences were observed in terms of overall and major complications between the prehabilitation and standard care groups. Pancreatic leak was not statistically different between the groups (11% vs 27%, p = 0.204). However, DGE was significantly lower in the prehabilitation group (5.6% vs 40.9% in the standard care group, p = 0.01). CONCLUSION: prehabilitation did not reduce postoperative complications following pancreaticoduodenectomy. However, a reduction in DGE was observed. Further studies are needed to validate the role and the timing of prehabilitation in high-risk patients.

A meta-analysis of randomized controlled trials comparing laparoscopic vs open pancreaticoduodenectomy.

BACKGROUND: Recent randomized controlled trials (RCTs) reported conflicting results regarding the safety of laparoscopic pancreaticoduodenectomy (LPD). The aim of this study was to perform a meta-analysis of the available RCTs concerning the short-term outcomes of LPD versus open pancreaticoduodenectomy (OPD). METHODS: The Cochrane Central Register of Controlled Trials, MEDLINE (through PubMed), EMBASE, Scopus databases and ClinicalTrials.gov register were searched. Only RCTs published up to February 2019 were eligible for inclusion. Random-effect models were used to summarize the relative risks (RR) and mean differences. RESULTS: 3 RCTs were identified, including a total number of 114 and 110 patients who underwent LPD and OPD, respectively. The rate of major postoperative complications (Clavien-Dindo ≥3) was 29% in LPD vs 31% in OPD group (RR 0.80 (95% CI: 0.36-1.79); p = 0.592). Complication-related mortality occurred in 5% (LPD) vs 4% (OPD) patients (RR 1.22 (95% CI: 0.19-8.02); p = 0.841). LPD was significantly associated with longer operative time [95 min (95% CI: 24-167; p = 0.009)] and lower perioperative blood loss [-151 ml (95% CI: 169-133; p < 0.001)]. CONCLUSIONS: There are no statistically significant differences between LPD and OPD in terms of postoperative complications and mortality. However, these findings should be interpreted with caution due to high clinical and statistical heterogeneity of pooled data. Further studies with different outcome measures are needed to clarify the future of LPD.

Enzymatic mediastinitis as a fearsome complication of pancreatic necrosectomy.

Several types of thoracic complications can be associated with severe acute pancreatitis. Some are very common, such as pleural effusion, whilst some others are very rare although life threatening, such as enzymatic mediastinitis (EM). Only a few cases of EM (when related to acute pancreatitis) have been described in the literature. Here we describe the case of a 76-year-old female who developed acute respiratory failure and atrial fibrillation during the postoperative period after an open pancreatic necrosectomy, due to an EM episode. The mediastinal collection was drained by thoracotomy, following an improvement of the patient's general condition. This is the first case of EM following surgical management for acute necrotizing pancreatitis. EM is a rare but life threatening complication that usually requires surgery.

Evidence-based Guidelines for the Management of Exocrine Pancreatic Insufficiency After Pancreatic Surgery.

OBJECTIVE: To provide evidence-based recommendations for the management of exocrine pancreatic insufficiency (EPI) after pancreatic surgery. BACKGROUND: EPI is a common complication after pancreatic surgery but there is certain confusion about its frequency, optimal methods of diagnosis, and when and how to treat these patients. METHODS: Eighteen multidisciplinary reviewers performed a systematic review on 10 predefined questions following the GRADE methodology. Six external expert referees reviewed the retrieved information. Members from Spanish Association of Pancreatology were invited to suggest modifications and voted for the quantification of agreement. RESULTS: These guidelines analyze the definition of EPI after pancreatic surgery, (one question), its frequency after specific techniques and underlying disease (four questions), its clinical consequences (one question), diagnosis (one question), when and how to treat postsurgical EPI (two questions) and its impact on the quality of life (one question). Eleven statements answering those 10 questions were provided: one (9.1%) was rated as a strong recommendation according to GRADE, three (27.3%) as moderate and seven (63.6%) as weak. All statements had strong agreement. CONCLUSIONS: EPI is a frequent but under-recognized complication of pancreatic surgery. These guidelines provide evidence-based recommendations for the definition, diagnosis, and management of EPI after pancreatic surgery.

Do some patients receive unnecessary parenteral nutrition after pancreatoduodenectomy? Results from an international multicentre study.

Backgrounds/Aims: After pancreatoduodenectomy (PD), an early oral diet is recommended; however, the postoperative nutritional management of PD patients is known to be highly variable, with some centers still routinely providing parenteral nutrition (PN). Some patients who receive PN experience clinically significant complications, underscoring its judicious use. Using a large cohort, this study aimed to determine the proportion of PD patients who received postoperative nutritional support (NS), describe the nature of this support, and investigate whether receiving PN correlated with adverse perioperative outcomes. Methods: Data were extracted from the Recurrence After Whipple's study, a retrospective multicenter study of PD outcomes. Results: In total, 1,323 patients (89%) had data on their postoperative NS status available. Of these, 45% received postoperative NS, which was "enteral only," "parenteral only," and "enteral and parenteral" in 44%, 35%, and 21% of cases, respectively. Body mass index < 18.5 kg/m2 (p = 0.03), absence of preoperative biliary stenting (p = 0.009), and serum albumin < 36 g/L (p = 0.009) all correlated with receiving postoperative NS. Among those who did not develop a serious postoperative complication, i.e., those who had a relatively uneventful recovery, 20% received PN. Conclusions: A considerable number of patients who had an uneventful recovery received PN. PN is not without risk, and should be reserved for those who are unable to take an oral diet. PD patients should undergo pre- and postoperative assessment by nutrition professionals to ensure they are managed appropriately, and to optimize perioperative outcomes.

Patterns, timing and predictors of recurrence following pancreaticoduodenectomy for distal cholangiocarcinoma: An international multicentre retrospective cohort study.

INTRODUCTION: Patients undergoing pancreaticoduodenectomy for distal cholangiocarcinoma (dCCA) often develop cancer recurrence. Establishing timing, patterns and risk factors for recurrence may help inform surveillance protocol strategies or select patients who could benefit from additional systemic or locoregional therapies. This multicentre retrospective cohort study aimed to determine timing, patterns, and predictive factors of recurrence following pancreaticoduodenectomy for dCCA. MATERIALS AND METHODS: Patients who underwent pancreaticoduodenectomy for dCCA between June 2012 and May 2015 with five years of follow-up were included. The primary outcome was recurrence pattern (none, local-only, distant-only or mixed local/distant). Data were collected on comorbidities, investigations, operation details, complications, histology, adjuvant and palliative therapies, recurrence-free and overall survival. Univariable tests and regression analyses investigated factors associated with recurrence. RESULTS: In the cohort of 198 patients, 129 (65%) developed recurrence: 30 (15%) developed local-only recurrence, 44 (22%) developed distant-only recurrence and 55 (28%) developed mixed pattern recurrence. The most common recurrence sites were local (49%), liver (24%) and lung (11%). 94% of patients who developed recurrence did so within three years of surgery. Predictors of recurrence on univariable analysis were cancer stage, R1 resection, lymph node metastases, perineural invasion, microvascular invasion and lymphatic invasion. Predictors of recurrence on multivariable analysis were female sex, venous resection, advancing histological stage and lymphatic invasion. CONCLUSION: Two thirds of patients have cancer recurrence following pancreaticoduodenectomy for dCCA, and most recur within three years of surgery. The commonest sites of recurrence are the pancreatic bed, liver and lung. Multiple histological features are associated with recurrence.