Left Pancreatectomy Research Articles

Case 200: Gastric Enterochromaffinlike Cell Tumors in a Patient with Type 1 Multiple Endocrine Neoplasia

History A 55-year-old man presented with chronic epigastric pain lasting for about 1 year and without fever or vomiting. The abdomen was soft and tender at physical examination. Laboratory tests revealed unremarkable liver function, normal hemoglobin level, and normal amylase level. White blood cell count was normal, and there was no inflammatory syndrome. The patient's medical history included pancreatic gastrinoma resected by means of left pancreatectomy 31 years before, hyperparathyroidism treated with subtotal parathyroidectomy 24 years before, and a slowly growing lung mass known for 9 years. Esophagogastroduodenoscopy was performed because of a suspected gastroduodenal ulcer. The results showed numerous small (<10 mm) gastric and duodenal ulcers and multiple 10-15-mm polypoid gastric masses. Contrast material-enhanced dual-phase multidetector row computed tomography (CT) of the chest and abdomen was performed with a 64-section CT scanner (LightSpeed VCT; GE Healthcare, Milwaukee, Wis). Technical parameters for CT were as follows: pitch, 0.98; section thickness and reconstruction interval, 1.25 mm; 120 kVp; and variable milliamperage determined by x-, y-, and z-axis dose modulation. After an unenhanced abdominal scan, iobitridol, a nonionic iodinated contrast agent containing 350 mg of iodine per milliliter (Xenetix 350; Guerbet, Aulnay-sousbois, France), was administered intravenously through a 16-18-gauge catheter. A 120-mL dose of the contrast agent was injected via an antecubital vein at a rate of 4 mL/sec. No oral contrast medium was administered. After preliminary unenhanced abdominal scanning, arterial and portal venous phase acquisitions were obtained 45 and 80 seconds after initiation of contrast medium injection.

Cost benefit analysis of mesh reinforcement of stapled left pancreatectomy

ObjectivesPancreatic leak is a morbid complication following left pancreatectomy, which results in prolonged hospitalization, additional diagnostic testing and invasive procedures. The present authors have previously demonstrated that mesh reinforcement of stapled left pancreatectomy results in fewer pancreatic leaks. This study was conducted to investigate whether mesh reinforcement also results in cost benefits for the health care system. MethodsA cost benefit model was developed to estimate net cost savings from the payer's perspective. The model is based on the results of a randomized, single-blinded trial of mesh versus no mesh reinforcement of the pancreatic remnant after left pancreatectomy. A two-way sensitivity analysis was conducted to determine the model's sensitivity to fluctuations in the cost of mesh and the effectiveness of the mesh in reducing clinically significant leaks. ResultsAverage total costs for an episode of care were US$13337 and US$15505 for patients who did and did not receive mesh, respectively, which indicates savings of US$2168. Two-way sensitivity analysis showed that, given a probability of 1.9% for developing a clinically significant leak in patients in whom mesh reinforcement was used, the strategy would continue to save costs if mesh were priced at ≤US$1804. ConclusionsMesh reinforcement decreases clinically significant pancreatic leaks. Despite the additional cost of mesh reinforcement, the use of mesh reinforcement results in overall cost savings for the health care system because of the resultant decrease in the occurrence of clinically significant leaks.

Pancreatic Resections: Are There Preoperative Factors Related to a “Soft Pancreas” and Are They Useful in Predicting Pancreatic Fistulas?

Context Soft pancreatic parenchyma is the most widely recognized risk factor for pancreatic fistulas. Objective To evaluate factors related to a “soft pancreas”. Methods Retrospective study of patients who underwent pancreaticoduodenectomy or left pancreatectomy. Factors considered were: sex, age, co-morbidities, body mass index, American Society of Anesthesiologists (ASA) score, characteristics of the pancreatic remnant and preoperative diagnosis. Principal endpoints were: to recognize preoperative factors related to a soft pancreatic remnant and to establish if they are useful in predicting a pancreatic fistula. Results Two hundred and eight patients were recorded. The pancreatic fistula rate was 29.3% and it was significantly related to the type of resection, the soft texture of the pancreatic remnant (P=0.037), a Wirsung duct diameter 24 kg/m 2 (P=0.011); a Wirsung duct size ≤3 (P 24 kg/m 2 , Wirsung duct size <3 mm and preoperative diagnosis may be considered as preoperative factors related to a soft pancreas and they resulted useful in predicting the incidence of pancreatic fistulas.

Primary Carcinoid Tumors of the Pancreas: Report of Eight Cases

Context Primary pancreatic carcinoid tumors (foregut) are very rare. A typical carcinoid syndrome with diarrhea and flushing may be present. The diagnosis is based on the high urinary 5-HIAA (5-hydroxyndole acetic acid) levels (or high serum serotonin levels) or the immunostaining of serotonin (5-HT) in the tumor cells. Objective We evaluated clinical presentation, endocrine tumor markers, histology, therapeutic approach and follow up. Methods From 1986 to 2011 in our department we observed 211 neuroendocrine (NE) pancreatic tumors and 8 of them (3.8%) were primary carcinoid tumors (5 males and 3 females; averaging 55.8 years, range: 38-69 years). Follow up was updated until December 2012. Results Among the eight patients enrolled, 3 were symptomatic. Seven had high serum 5-HT or high urinary 5-HIAA, and one was asymptomatic with immunostaining of 5-HT in tumor cells. Location: 6 body-tail. All were malignant tumors: 7 liver and 1 single nodal metastases. Markers: 4 high serum 5-HT (up to 176 µmol/L), 7 high urinary 5-HIAA (up to 522 µmol/L). Surgery: 1 left pancreatectomy, 7 biopsy. Histology: 7 NE tumor, 1 negative pancreatic biopsy (liver metastases). Other therapy: 3 treated with somatostatin analogues (SST-A) and chemotherapy (CT), 1 CT and radiometabolic therapy after hepatic artery embolization (HAE), 1 HAE and SST-A, 1 CT. Follow up: 6 dead for disease progression (mean survival 52 months), 2 alive (1 without disease 78 months after surgery; 1 asymptomatic with high 5-HIAA 33 months after SST-A and CT). Conclusion Most of primary pancreatic carcinoids are locally advanced tumors or have liver metastases at time of diagnosis, then patients are not amenable to surgery. Although most patients had high 5-HIAA urinary excretion, few patients had carcinoid syndrome. A long term survival may be achieved with multimodal approach, including chemotherapy, in foregut carcinoid tumors.

Laparoscopic Left Pancreatectomy for Primary Pancreatic Adenocarcinoma

Context Laparoscopic left pancreatectomy has been shown to be feasible, safe and efficient technique for the treatment of benign and pre-malignant lesions of the left half side of the pancreas, but there is still controversy when it comes to the management of the malignant lesion of the pancreas. Objective To evaluate the feasibility, outcomes and oncological efficiency of the laparoscopic approach of primary pancreatic adenocarcinoma in two major tertiary centers. Methods Data from a prospectively updated databases of 105 patients operated between 2005 and 2013 was analyzed. Data included demographics, conversion rate, operative time, blood loss, length of hospital stay, postoperative complications according to Clavien-Dindo classification (score ≥3), mortality, lymph node retrieval and margin status. Results Fifteen patients (14.3%) underwent laparoscopic left pancreatectomy (LLP) for ductal adenocarcinoma; 7 patients were (46.7%) females and 8 (53.3%) males. The mean age of the patients was 62 years (range: 40-74 years). In three patients a multivisceral en bloc resection was accomplished (adrenalectomy: 2; mesocolonic resection: 3; and duodenojejunal flexure: 2). One patient was converted (6.6%). Mean operative time was 230 min, (range: 120-370 min). Mean blood loss was 228 mL (range: 50-900 mL). Average post-operative stay was 6 days (range: 2-17 days). Complication rate according to the classification of Clavien-Dindo (score ≥3) was 13.3% (2/15) and grade B/C postoperative pancreatic fistula rate was 20% (3/15). Average nodal rate was 14 (range: 3-41). R0 resection was achieved in 85.7% (12/14; in one patient R status is not yet available). Conclusion In expert centers, laparoscopic left pancreatectomy for primary pancreatic malignancy is feasible with excellent short term clinical and oncological results. However randomized studies are not yet available and are now seriously needed to shed light on future direction.

Extending Indications for Islet Autotransplantation in Pancreatic Surgery

To assess metabolic and oncologic outcomes of islet autotransplantation (IAT) in patients undergoing pancreatic surgery for either benign or malignant disease. IAT is performed to improve glycemic control after extended pancreatectomy, almost exclusively in patients with chronic pancreatitis. Limited experience is available for other indications or in patients with pancreatic malignancy. In addition to chronic pancreatitis, indications for IAT were grade C pancreatic fistula (treated with completion or left pancreatectomy, as indicated); total pancreatectomy as an alternative to high-risk anastomosis during pancreaticoduodenectomy; and distal pancreatectomy for benign/borderline neoplasm of pancreatic body-neck. Malignancy was not an exclusion criterion. Metabolic and oncologic follow-up is presented. From November 2008 to June 2012, 41 patients were candidates to IAT (accounting for 7.5% of all pancreatic resections). Seven of 41 did not receive transplantation for inadequate islet mass (4 pts), patient instability (2 pts), or contamination of islet culture (1 pt). IAT-related complications occurred in 8 pts (23.5%): 4 bleeding, 3 portal thromboses (1 complete, 2 partial), and 1 sepsis. Median follow-up was 546 days. Fifteen of 34 patients (44%) reached insulin independence, 16 patients (47%) had partial graft function, 2 patients (6%) had primary graft nonfunction, and 1 patient (3%) had early graft loss. Seventeen IAT recipients had malignancy (pancreatic or periampullary adenocarcinoma in 14). Two of them had already liver metastases at surgery, 13 were disease-free at last follow-up, and none of 2 patients with tumor recurrence developed metastases in the transplantation site. Although larger data are needed to definitely exclude the risk of disease dissemination, the present study suggests that IAT indications can be extended to selected patients with neoplasm.

Case 200

Case 200

Laparoscopic left pancreatectomy for pancreatic sarcomatoid carcinoma: A case report and review of the literature

Sarcomatoid carcinoma of the pancreas is extremely rare. The current report presents a case of carcinosarcoma of the pancreas in a 48-year-old male. Pre-operative computed tomography scans revealed a large complex cystic and solid mass in the tail of the pancreas; the patient underwent a laparoscopic spleen-preserving left pancreatectomy. The tumor was shown to be made of cystic and solid components, with a grossly grey/ white appearance. A histological evaluation of the tumor revealed two elements separated from each other, one component was a pancreatic ductal adenocarcinoma and the other component exhibited a sarcomatous growth pattern, composed of spindle cells and multinucleated giant cells. Immunohistochemically, the epithelial area was positive for cytokeratin (CK) and negative for vimentin, while the sarcomatoid area was negative for CK and positive for vimentin. These observations confirmed a diagnosis of pancreatic carcinosarcoma. Although the patient was treated by gemcitabine following surgery, the outcome was extremely poor and the patient succumbed to sarcomatoid carcinoma three months after the treatment.

P-0172 - Surgical Treatment of Malignant Nonfunctioning Neuroendocrine Tumors of the Pancreas

Background: Surgical technique of treatment of these tumors may be similar to the treatment of adenocarcinoma or organ preserving technique is preferable in case when diagnosis of neuroendocrine tumors was confirmed in preoperative period. Neuroendocrine tumors of the pancreas have a more favorable course and prognosis comparing to pancreatic adenocarcinoma and occupy approximately 1-2% of all oncologic pathology of the pancreas. Aims: Assessment of the efficacy, safety and long-term outcome of treatment of patients with non-functioning malignant neuroendocrine tumors (MNET) of pancreas. Methods: From 2004 to 2012 we had treated 87 patients with nonfunctioning neuroendocrine tumors of the pancreas, in 65 (74.7%) patients had malignant tumors, neuroendocrine tumor G2 - 43 (66.2%) patients and neuroendocrine carcinoma G3 - 22 (33.8%) (WHO classification, 2010). Localization of MNET were head of the pancreas in 27 (41.5%) patients, in the body - 15 (23.1%), in the tail - 19 (29.2%), in the uncinate pancreas - 4 (6.2%). Before to surgery, the diagnosis of neuroendocrine tumor of the pancreas was verified in 12 (18.4%) patients, intraoperatively in 8 (12.3%) patients. In all cases, the diagnosis was confirmed by histological and imunohistochemical researches. It should also be noted that, in further study on neuroendocrine tumors were re-analyzed by histological specimens of patients with 3-year disease-free survival in patients with intraductal adenocarcinoma, for 4 patients a diagnosis was changed retrospectively on malignant neuroendocrine tumor of the pancreas (3 patients - G2 and 1 patient - G3 MNET). Results: Pancreaticoduodenectomy (PDE) (Whiple procedure) had performed for 19 (29.2%) patients, 6 (9.2%) patients pylorus preserving PDE, in 2 (3.1%) cases - PDE and liver resection, total pancreatectomy in 1 (1.5%) cases, left side pancreatectomy with splenectomy - 16 (24.6%), spleen-preserving left side pancreatectomy - 7 (10.8%), laparoscopic left side pancreatectomy with splenectomy - 4 (6.2%), central pancreatectomy - 3 (4,6%), palliative operation - 3 (4,6%), explorative laparotomy 1 (1.5%). Postoperative complications were happened with 24.6% of patients, 3 patients (4.6%) had postoperative acute pancreatitis, 12 patients (18.5%) had pancreatic fistula (grade A - 8 patients, Grade B - 3 patients, Grade C - 1 patient), and 1 (1.5%) patient had intra-abdominal bleeding. The overall 5-years survival without metastasis was 67.4%, in case of the metastasis presence only into the regional lymph nodes - 52.7%, when metastases were detected in the liver 5-years survival was 14.2%. The disease-free survival in patients with the absence of distant metastases was 42.6 months and 11.2 with distant metastases. Conclusion: Laparoscopic surgery and organ-preserving surgery are safe and effective operations for local malignant nonfuctioning neuroendocrine tumors. The presence of distant metastases of malignant neuroendocrine tumors of the pancreas decrease the long-term outcome. But are not contraindications for multivisceral resection surgery and give a better long-term outcome compared to palliative surgery. The presence of metastases in regional lymph nodes were not significantly impair long-term outcome.

Middle-preserving pancreatectomy: report of two cases and review of the literature

BackgroundMiddle-preserving pancreatectomy (MPP) is a parenchyma-sparing surgical procedure which has recently been sporadically reported for the treatment of multicentric periampullary-pancreatic lesions. However, a comprehensive recognition of this procedure has not been clearly elucidated.Case presentationWe herein report two patients undergoing MPP due to synchronous multicentric pancreatic neoplasm. Patient one was a 24-year-old woman with a multicentric solid pseudopapillary neoplasm (SPN) and patient two was a 36-year-old woman with a multicentric serous cystic neoplasm (SCN). Simultaneous atypical pancreaticoduodenectomy and atypical left pancreatectomy were performed in patient one; simultaneous standard pancreaticoduodenectomy and atypical left pancreatectomy with spleen preservation were performed in patient two. Approximately 6 cm and 5 cm segments of the middle portion of the pancreas were preserved, respectively. At follow-up at 36 months and 6 months respectively, patient one had developed diabetes and malabsorption requiring dietary control, exercise and pancreatic enzyme supplement whereas patient two showed normal fasting blood glucose without diarrhea. Both patients were disease-free and in good nutritional condition. We reviewed twenty cases of MPP that were previously reported in the literature. Patient characteristics, surgical techniques and short- and long-term outcomes were analyzed.ConclusionMPP is mainly beneficial for multicentric noninvasive periampullary-pancreatic lesions. However, for multicentric periampullary-pancreatic lesions involving even primary invasive cancers, as long as the invasive cancers affect only one side of the pancreas (proximal or distal), MPP could serve as a rational choice in well-selected patients.

Outcomes of pancreatic resection: Analysis of the french national database on 6436 patients

Outcomes of pancreatic resection: Analysis of the french national database on 6436 patients

Surgical duodenal ampullectomy in the treatment of ampullary neoplasm: 12 cases experience

Surgical duodenal ampullectomy in the treatment of ampullary neoplasm: 12 cases experience

Laparoscopic Distal Pancreatectomy with Preservation of the Spleen and of the Splenic Vessels, with Autologous Islet Transplantation. a New Minimally Invasive Technique to Treat Benign Neoplasm of the Pancreatic Body. Case Series of 3 Patients

Introduction: Benign pancreatic neoplasm located in the left pancreas can require resection because of the risk of malignant transformation. Distal pancreatectomy for body-neck lesions frequently results in postoperative diabetes because of the great amount of normal resected pancreatic tissue. Autologous islet transplantation (AIT) can prevent this event. The association of AIT, laparoscopic technique and the preservation of the spleen can minimize all the possible consequences of surgery in such cases. We herein present a case series of 3 non diabetic patients who underwent spleen preserving left laparoscopic pancreatectomy for symptomatic cystic lesions. Materials and methods: Three non-diabetic, female patients (age 37-44) were candidate to surgery because of pancreatic cystic lesions between May and September 2011. The lesions were located at the body/neck of the pancreas and measured 3, 4 and 10cm respectively. The planned operation was a laparoscopic spleen-preserving left pancreatectomy with preservation of the splenic vessels, associated with AIT from the normal pancreas distal to the neoplasm. Results: The procedure was successfully performed in all the three cases. Mean operation time was 266 minutes. Surgical complications occurred in one patient (grade A pancreatic fistula). Postoperative stay was 6, 6, and 7. Histopathological evaluation revealed mucinous cystic neoplasm in 2 cases e serous cystic neoplasm in 1 case. Pancreas weight was 33, 22, 30g and 105.200, 40.390, 94.790 Islet equivalents were isolated respectively and infused in the portal vein by percutaneous transhepatic approach the day after surgery. No early postoperative insulin administration was needed. One patient developed a transient partial portal thrombosis 2 months after islet infusion. The 3 patients are insulin independent at a mean follow up of 7 months. Conclusions: The combination of minimally invasive approach and autologous islets transplantation is safe and effective in prevent post surgical diabetes after resection for benign neoplasm located in the body-neck of the pancreas.

Short-term outcome and cost-benefit analysis of laparoscopic left pancreatectomy: Results of a case-match study

Short-term outcome and cost-benefit analysis of laparoscopic left pancreatectomy: Results of a case-match study

A prospective non-randomised study, comparing Frey procedure with pylorus-preserving pancreatoduodenectomy in the treatment of chronic pancreatitis

s / Pancreatology 12 (2012) 502–597 522 Objectives: Current applications of laparoscopic approaches to left pancreatectomy, tumor enucleation, and pancreaticoduodenectomy for treatment of pancreatic tumors are considered in the available evidence demonstrating feasible and safety. Patients: Laparoscopic distal pancreatectomy (LDP) 5cases (MCN3, IPMN1, Insulinoma1), pancreaticoduodenedctomy (LPD) 1case (Cancer of the Papilla), and tumor enucleation 1case (Insulinoma) are performed. Result: (1) MCN: The mean age of the 3 patients was 34.3 years, and were all women. The mean operative time, blood loss and hospital stay were 330min, 122ml, 8.6days. (2)IPMN: A 76-years-old women had the F40mm tumor of the tail of the pancreas. The operative time, blood loss and hospital stay were 295min, 100ml, 8days. (3) Insulinoma: LPD; A 63years-old man had the F15mm tumor of the body of the pancreas. The operative time, blood loss and hospital stay were 440min, 200ml, 6days. Tumor enucleation; A 27-years-old man had the F17mm tumor of the body of the pancreas. The operative time, blood loss and hospital stay were 202min, A little, 7days. (4) Cancer of the Papilla: A 72-years-old woman had the T2 tumor with no involvement of lymph nodes. The operative time, blood loss and hospital stay were 558min, 70ml, 20days. Conclusion: Laparoscopic pancreatic surgery can be performed safely and effectively in patients with benign or low-grade malignant neoplasms of the pancreas.

A Comparison of Surgical Outcome of Major Pancreatic Resections: Left Pancreatectomy Versus Pancreaticoduodenectomy

Context Left pancreatectomy (LP) is considered a minor operation, when compared to pancreatico­duodenectomy (PD), due to a technically easier resection, without the burden of pancreatic anastomosis. Nonetheless LP is a demanding procedure with high blood loss and a high rate of postoperative pancreatic fistula. Objective To compare surgical outcome between patients undergoing PD and LP in a high volume institution. Methods Retrospective study on prospectively collected data on 981 consecutive patients undergoing PD (n=664) and LP (n=317), between 2004 and 2011. Postoperative complications were stratified according to Clavien-Dindo classifications (major complications III-V). Pancreatic fistula (PF) was defined according to ISGPF classifications. Results Mortality was higher in PD (3.6% vs . 0.3%, P<0.001), but overall morbidity was the same (63% in both groups). Postoperative major complications were more frequently in PD group (18.4% vs . 5.3%, P<0.001). PF was more frequent in LP (PD 30% vs . LP 46%, P<0.001); however, clinically significant grade B and C PF rate was higher in PD (16% vs . 6%, P<0.001). Intraoperative blood loss was similar between groups (PD 568±388 mL, LP 619±685 mL, P=0.169), however the rate of patients who received a blood transfusion intraoperatively was higher in the PD group (35% vs . 22%, P<0.001). This was probably due to a higher number of patients presenting with preoperative anemia (mean hemoglobin: PD 12.4±2.2 g/dL vs . LP 13.2±1.7 g/dL, P<0.001). PD patients were transfused more postoperatively (20% vs . 11%, P<0.001) and had a higher mean volume of postoperative transfusion (232±811 mL vs . 79±281 mL, P<0.001). Conclusion As expected, mortality and major complications were higher after PD. Overall morbidity and intraoperative blood loss were similar between the two operations.

Are There Predicting Factors Related to a “Soft Pancreas”?

Context It is well-known that a soft pancreatic stump is related to an increased incidence of postoperative pancreatic fistula (POPF) after pancreatic resection. Objectives To evaluate the presence of preoperative factors predicting soft pancreas in patients who underwent pancreaticoduodenectomy (PD) or left pancreatectomy (LP). Methods From January 2004 to June 2012, data regarding 208 consecutive patients whom underwent PD or LP were collected in a prospective database. For each patients we recorded sex, age, co-morbidities, BMI, ASA score, preoperative diagnosis, type of resection, characteristics of pancreatic remnant, pathological diagnosis and POPF rate. Univariate and multivariate analyses were carried out in order to evaluate the preoperative factors predicting a soft pancreatic stump. Results There were 102 (49%) female and 106 (51%) male with a mean age of 64.5±13.2 years. Co-morbidities were present in 128 (61.5%) patients; 64 patients (30.8%) were ASA II, 128 (61.5%) ASA III and 16 (7.7%) ASA IV. Mean BMI was 25.3±4.4 kg/m 2 . A pancreatic cancer or a chronic pancreatitis (CP) were suspected in 85 (40.9%) cases. One-hundred and twenty-five (60.1%) patients underwent PD and 83 (39.9%) LP. Pancreatic stump was soft in 129 (62%) cases and Wirsung duct was dilated in 55 cases (26.4%). Sixty-one patients (29.3%) had POPF (9.1% grade A; 19.2% grade B; and 1% grade C). Univariate analysis showed that a non dilated Wirsung duct (P 24 kg/m 2 , a non dilated Wirsung duct and a pancreatic lesion different from pancreatic cancer or CP can predict a soft pancreas and subsequently a major risk of POPF.

Reactive C Protein and White Blood Cells Levels as Early Predictors of Postoperative Inflammatory Complications After Pancreatic Surgery

Context Pancreatic surgery is challenging and associated with high morbidity, even in high-volume surgical settings; this is probably the reason why, while in other surgical areas fast-track recovery programs are nowadays widely performed, pancreatic surgeons are still reluctant in applying them. The identification of clinical and diagnostic criteria to early predict postoperative inflammatory complications (PIC) development could be useful in tailoring postoperative management to patient personal risk. Objective Aim of the study is the assessment of diagnostic accuracy of reactive C protein (RCP) and white blood cells (WBC) levels as early predictors of PIC in pancreatic surgery. Methods Between January 2010 and June 2012 we performed 225 pancreatic resections for benign and malignant disease, of which: 136 pancreatico­duodenectomies (PD), 65 left pancreatectomies (LP), 11 total pancreatectomies (TP), 12 enucleations (EN). Postoperative levels of RCP (detected by our laboratory with a high-sensitivity method) and WBC from postoperative day 1 (POD 1) to POD 7, recorded in our perspective electronic database, were analyzed searching for association with PIC (anastomotic leakage, sepsis, airways, urinary tract and wound infection, abdominal collection); using the receiver operating characteristic method (ROC curve), diagnostic accuracy was evaluated by an area under the curve (AUC) analysis. Results PIC occurred in 39.6% of patients (43.4% after PD; 30.3% after LP; 45.5% after TP, 41.7% after EN). Cancer diagnosis, preoperative chemotherapy, age and ASA score did not influence PIC rates. Mean RCP levels were significantly higher in patients who developed PIC each day from POD 1 to POD 7 (P<0.001), while mean WBC levels were significantly higher in this group only from POD 4 to POD 7 (P<0.001). The highest diagnostic accuracy was observed for RCP levels on POD 4 (AUC=0.835), with a cut-off value of 14,62 mg/L, whose sensibility and specificity were 83% and 81%, respectively. In a similar way, RCP postoperative levels resulted significantly associated, from POD 1 to POD 7, with high grade PIC (grade II-V according to Clavien-Dindo classification) (P<0.001). We could not identify any valuable cut-off for WBC. Conclusion RCP postoperative level on POD 4 appears to be an useful early predictor for PIC in pancreatic surgery and could guide patient’s management (fast track recovery programs and/or further diagnostic research for septic processes); otherwise, WBC values, probably more influenced than RCP by physiological postoperative acute inflammatory response, fail in decisively distinguishing patients who are developing PIC.

Laparoscopic Versus Open Left Pancreatectomy: Short Term Outcome and Cost-Benefit Analysis

Context An increasing number of surgeons are today performing laparoscopic left pancreatectomy (LLP), since available nonrandomized studies demonstrated its feasibility, safety and oncologic adequacy. However, most existing data come from small single-institution reports or from heterogeneously composed multicenter comparisons. Moreover, there is very limited information about economic implications of minimally invasive pancreatic surgery. Objective This study reports our experience in laparoscopic left pancreatectomy compared with open technique (OLP), assessing perioperative outcomes and financial impact of this procedure in a high volume surgical setting. Methods Between February 2009 and June 2011 we performed 112 left pancreatectomies, 53 of which (47%) were LLP. Excluding the initial learning curve, the remaining 43 patients were matched with a control group selected from our perspective electronic database. Match criteria were gender, age, ASA score, BMI, lesion site, malignant or benign disease. Results Mean operative time was similar (LLP 216±61 min; OLP 214±7 min; P=0.885), blood loss was reduced in LLP (388±371 mL vs. 571±599 mL, P=0.092), especially in cancer patients (514±350 mL vs . 946±787 mL, P=0.072); intraoperative transfusion and unplanned splenectomy rates were similar. Larger lesions were associated with increased unplanned splenectomy rate. Conversion rate (CR) was 18%; higher BMI (>30 kg/m 2 ) and pancreatic body site were associated with increased CR. There were no differences in positive margin rates, number of nodes examined and number of N1 patients. There was no mortality in both groups. Overall morbidity was equable (63% in LLP, 60% in OLP; P=0.958), as well as major complication rate (7% in LLP, 3% in OLP; P=0.604). Clinically significant pancreatic fistula rate was 14% in LLP and 9% in OLP (P=0.728). No grade C fistulas were observed. An equal proportion of patients in each group was discharged before removing surgical drain. Delayed gastric empting, wound and urinary tract infection were more frequent in OLP. Mean LOS was 8.37 days in LLP vs . 8.81 days in OLP (P=0.481); LOS in non complicated patients was 6.96 days in LLP vs . 7.50 days in OLP (P=0.220). Mean number of diagnostic test, transfusion rate, antibiotic administration and readmission rate were similar. Each patient of LLP group saved €168.47 because of shorter LOS and slightly fewer complication cost, generating however an extra cost of €767,01 due to more expensive surgical instruments. Conclusion This study confirms safety and oncologic adequacy of this technique, identifying probable risk factors for conversion and demonstrating economic sustainability of LLP. Final balance still have to be realized considering indirect costs as shorter home convalescence, quality of life and better cosmetic result.

Post-Pancreatectomy Hemorrhage: Management, Outcome and Predicting Factors of a Life-Threatening Complication

Context Hemorrhagic complications are life-threatening complications occurring after pancreatectomy. Objectives 1. To analyze incidence, management, and clinical outcome of post-pancreatectomy hemorrhage (PPH). 2. To identify factors associated with its occurrence. Methods Retrospective study on prospectively collected data about 981 consecutive patients undergoing a standard pancreatic resection (pancreaticoduodenectomy or left pancreatectomy) between 2004 and 2011. PPH was defined as postoperative clinical or radiological evidence of bleeding, requiring transfusion of packed red blood cells. PPH was stratified based on: time of onset (early: within 24 hr after surgery; late >24 hr); site (intraluminal: in the GI tract; or extraluminal); severity (mild: self-limiting with no need for radiologic or surgical intervention; severe: requiring intervention). Results A total of 65 patients (6.6%) experienced PPH. It was early in 17 (26%) and late in 48 (74%) patients. In 9 cases (14%) PPH was intraluminal, in 56 (86%) extraluminal. In 14 patients (21%) PPH was mild; angiography (with embolization or stenting) was used in 9 patients (14%), and it was successful in 5 patients; the remaining 4 patients needed further radiological or surgical treatment. Relaparotomy for PPH was necessary in overall 49 patients (75%). Mean volume of postoperative transfusion was 1,847±1,997 mL. Re-bleeding after radiological or surgical intervention occurred in 10 patients (15%). Overall mortality in patients with PPH was 45%. Factors predicting PPH were type of surgery (pancreaticoduodenectomy), soft pancreatic stump, and postoperative pancreatic fistula. Conclusion PPH can be a dramatic complication, frequently leading to relaparotomy, and carries a high risk of mortality.