Dissertationen zum Thema „Pancreatic resection“

Background: Metastases to the pancreas are rare, accounting for less then 2 % of all pancreatic malignancies. However, both the benefit of extended tumor resection and the ideal oncological approach have not been established for such cases; therefore, we evaluated patients with metastasis to the pancreas who underwent pancreatic resection. Methods: Between 1994 and 2012, 676 patients underwent pancreatic surgery in our institution. We retrospectively reviewed patients’ medical records according to survival, and surgical and non-surgical complications. Student’s t-test and the log-rank test were used for statistical analysis. Results: Eighteen patients (2.7 %) received resection for pancreatic metastases (12 multivisceral resections and 6 standard resections). The pancreatic metastases originated from renal cell carcinoma (n = 10), malignant melanoma (n = 2), neuroendocrine tumor of the ileum (n = 1), sarcoma (n = 1), colon cancer (n = 1), gallbladder cancer (n = 1), gastrointestinal stromal tumor (n = 1), and non-small cell lung cancer (n = 1). The median time between primary malignancy resection to metastasectomy was 83 months (range, 0–228 months). Minor surgical complications (Grade I-IIIa) occurred in six patients (33.3 %) whereas major surgical complications (Grade IIIb-V) occurred in three patients (16.6 %). No patients died during hospitalization. The median follow-up was 76 months (range, 10–165 months). One-year, 3-year and 5-year survival for standard resection versus multivisceral resection was 83, 50, and 56 % versus 83, 66, and 50, respectively. Twelve patients died after a median of 26 months (range, 5–55 months). Conclusions: A surgical approach with curative intent is justified in select patients suffering from metastases to the pancreas and offers good long-term survival. The resection of pancreatic metastases of different tumor types was associated with favorable morbidity and mortality when compared with resection of the primary pancreatic malignancies. Our findings also demonstrated that multivisceral resection was feasible, with acceptable long term outcomes, even though morbidity rates tended to be higher after multivisceral resection than after standard resection.

Background: Surgery remains the only curative option for the treatment of pancreatic adenocarcinoma (PDAC). The goal of this study was to investigate the clinical outcome and prognostic factors in patients after resection for ductal adenocarcinoma of the pancreatic head. Methods: The data from 195 patients who underwent pancreatic head resection for PDAC between 1993 and 2011 in our center were retrospectively analyzed. The prognostic factors for survival after operation were evaluated using multivariate analysis. Results: The head resection surgeries included 69.7% pylorus-preserving pancreatoduodenectomies (PPPD) and 30.3% standard Kausch-Whipple pancreatoduodenectomies (Whipple). The overall mortality after pancreatoduodenectomy (PD) was 4.1%, and the overall morbidity was 42%. The actuarial 3- and 5-year survival rates were 31.5% (95% CI, 25.04%-39.6%) and 11.86% (95% CI, 7.38%-19.0%), respectively. Univariate analyses demonstrated that elevated CEA (p = 0.002) and elevated CA 19–9 (p = 0.026) levels, tumor grade (p = 0.001) and hard texture of the pancreatic gland (p = 0.017) were significant predictors of a poor survival. However, only CEA >3 ng/ml (p < 0.005) and tumor grade 3 (p = 0.027) were validated as significant predictors of survival in multivariate analysis. Conclusions: Our results suggest that tumor marker levels and tumor grade are significant predictors of poor survival for patients with pancreatic head cancer. Furthermore, hard texture of the pancreatic gland appears to be associated with poor survival.

Background: The systemic immune-inflammation index based on peripheral neutrophil, lymphocyte, and platelet counts has shown a prognostic impact in several malignancies. The aim of this study was to determine the prognostic role of systemic immune-inflammation index in patients with pancreatic ductal adenocarcinoma undergoing resection. Methods: Consecutive patients who underwent surgical resection at the department of surgery at the Medical University of Vienna between 1995 and 2014 were included into this study. The systemic immune-inflammation index was calculated by the formula platelet*neutrophil/lymphocyte. Optimal cutoffs were determined using Youden's index. Uni-and multivariate analyses were calculated by the Cox proportional hazard regression model for overall survival. Results Three hundred twenty-one patients were included in this study. Clinical data was achieved from a prospective patient database. In univariate survival analysis, elevated systemic immune-inflammation index was found to be significantly associated with shortened patients' overall survival (p = 0.007). In multivariate survival analysis, systemic immune-inflammation index remained an independent prognostic factor for overall survival (p = 0.004). No statistical significance could be found for platelet to lymphocyte ratio and neutrophil to lymphocyte ratio in multivariate analysis. Furthermore, area under the curve analysis showed a higher prognostic significance for systemic immune-inflammation index, compared to platelet to lymphocyte ratio and neutrophil to lymphocyte ratio. Conclusion A high systemic immune-inflammation index is an independent, preoperative available prognostic factor in patients with resectable pancreatic ductal adenocarcinoma and is superior to platelet to lymphocyte ratio and neutrophil to lymphocyte ratio for predicting overall survival in pancreatic ductal adenocarcinoma patients.

Pancreatic (PC) is the eleventh most common malignancy in the UK but it has the poorest prognosis of all human adenocarcinoma. The autopsy, epidemiological and clinical studies have consistently identified PC as one of the most highly angiogenic and invasive malignancies, with the greatest prevalence and incidence of thrombo-embolism (TE). The incidence of TE in PC has been reported as high as 57%. Tissue factor (TF) bearing microparticles (MP) have recently been shown to promote thrombosis. The biological link between cancer haemostasis, cell invasion and angiogenesis remains unclear. These three indices may be driven by PC cells directly, be a reflection of the individual tumour stromal microenvironment and/or a result of the inflammatory response of the host. The hypothesis of the thesis is that factors directly attributed to the cancer promote the observed pathophysiology and that the removal of the tumour should result in reversal of these abnormalities. Flow cytometry was used for the evaluation of MP in plasma and quantification of surface-expressed TF, VEGFR-1 and-2 and EGFR. Cellular TF activity and pro-coagulant activity (prothrombin time) of PC patients were measured using a coagulometer. Matrigel Invasion Chambers and Boyden chambers with collagen IV were used to measure cellular invasion. A two dimensional angiogenesis assay was used to evaluate tubule formation in vitro in response to PC sera. Relative levels of protein expression of 55 angiogenic markers in the sera of PC patients were evaluated using a human angiogenesis array kit. Enzyme-linked immunosorbent assay was undertaken on VEGF, TF, TFPI, Leptin and annexin autoantibodies using sera or plasma from PC patients as appropriate. Immunohistochemical analysis of key markers of angiogenesis and thrombosis was also undertaken on resected PC samples. The in vitro optimisation experiments revealed that the cell invasion was significantly correlated with TF antigen expression and activity on PC cell lines (MIA-PaCa-2, AsPC-1 and CFPAC1) and that blocking TF on these cells decreased cell invasion. In the same manner neutralising soluble TF in PC serum samples also significantly decreased cell invasion, as did spiking of the serum with low molecular weight heparin. Analysis of sera from patients showed that TF bearing MP, pro-coagulant activity, cell invasion and angiogenesis (total length and number of capillaries) of PC cases were significantly higher than the control. Furthermore, the post-operative median number of TF bearing MP, procoagulant activity, cell invasion and angiogenesis (total length and total number of capillaries) were all significantly lower compared with pre-operative samples. Out of 55 angiogenic markers studied in 6 PC patients, pre- and post-operatively there was a significant decrease of angiopoietin-1, angiostatin/plasminogen, PDGF-AA, PDGF-AB/PDGF-BB and VEGF post-operatively. This result was supported by ELISA analysis of 29 samples and 14 controls that also showed significantly higher levels of VEGF in pancreatic cancer sera versus control groups, and that there was a significant decrease observed post-operatively only in the cancer patients. Furthermore both angiogenesis array and ELISA showed increased leptin levels post-operatively. Immunohistochemical analysis of the pancreatic tissue sections revealed that TF was expressed on 62 % of PC samples. There was significant correlation between TF expression on the tissue and procoagulant activity. Also, there was a significant correlation between tissue-expressed TF and in vitro angiogenesis, i.e. total length and number of capillaries. Furthermore, there was a significant correlation between TF expression on tissue with intratumoural microvascular density (MVD) and tumour-expressed with VEGFR 2. As expected, high levels of MVD correlated with high levels of tissue-expressed VEGF. Finally serum from patients who showed a high level of tissue-expressed VEGF also induced the greatest level of in vitro angiogenesis, i.e. number of capillaries. In summary, it was shown that TF expression on cell lines was significantly correlated with TF activity and cell invasion, and that TF expression in plasma and on tissue from PC patients was significantly correlated with procoagulant activity, cell invasion and angiogenesis. PC tissue-expressed VEGF was significantly associated with the angiogenic activity of PC sera and tissue MVD. Thus, the pathophysiology represented by a high procoagulant state, elevated cell invasion and angiogenic properties seen in PC patient sera appears to be driven by the malignant cells, as removal of the tumour causes a return towards the normal state.

Background Pancreatic heterotopia (PH) is defined as ectopic pancreatic tissue outside the normal pancreas and its vasculature and duct system. Most frequently, PH is detected incidentally by histopathological examination. The aim of the present study was to analyze a large single-center series of duodenal PH with respect to the clinical presentation. Methods A prospective pancreatic database was retrospectively analyzed for cases of PH of the duodenum. All pancreatic and duodenal resections performed between January 2000 and October 2015 were included and screened for histopathologically proven duodenal PH. PH was classified according to Heinrich’s classification (Type I acini, ducts, and islet cells; Type II acini and ducts; Type III only ducts). Results A total of 1274 pancreatic and duodenal resections were performed within the study period, and 67 cases of PH (5.3%) were identified. The respective patients were predominantly male (72%) and either underwent pancreatoduodenectomy (n = 60); a limited pancreas resection with partial duodenal resection (n = 4); distal pancreatectomy with partial duodenal resection (n = 1); total pancreatectomy (n = 1); or enucleation (n = 1). Whereas 65 patients (83.5%) were asymptomatic, 11 patients (18.4%) presented with symptoms related to PH (most frequently with abdominal pain [72%] and duodenal obstruction [55%]). Of those, seven patients (63.6%) had chronic pancreatitis in the heterotopic pancreas. The risk of malignant transformation into adenocarcinoma was 2.9%. Conclusions PH is found in approximately 5% of pancreatic or duodenal resections and is generally asymptomatic. Chronic pancreatitis is not uncommon in heterotopic pancreatic tissue, and even there is a risk of malignant transformation. PH should be considered for the differential diagnosis of duodenal lesions and surgery should be considered, especially in symptomatic cases.

Background: Pancreatic métastases from renal cell carcinoma (RCC) are clinically rare but highly resectable. The aim of this article is to identify patients who profit from pancreatic resection of RCC despite the invasiveness of the surgery. Methods: Between January 1996 and December 2007, data from 744 patients were collected in a prospective pancreatic surgery database, and patients with metastasis into the pancreas from RCC were identified. Results: Resective surgery was performed in 14 patients with metastasis to the pancreas from RCC. Most patients were clinically asymptomatic. The median interval between primary treatment of RCC and occurrence of pancreatic metastasis was 94 months (range 32–158). The morbidity rate was 42.8%. Patients with a metastasis size <2.5 cm had a much better survival after resection (100 months) than those with a metastasis size >2.5 cm (44 months). Moreover, the number of métastases predicts the survival after resection. Conclusions: In patients with pancreatic métastases from RCC who have only limited disease, complete resection of all lesions can be successfully performed with a low rate of complications. Thus, patients with a history of RCC should be monitored for more than 10 years after nephrectomy to detect recurrence
Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich

Background: Pancreatic métastases from renal cell carcinoma (RCC) are clinically rare but highly resectable. The aim of this article is to identify patients who profit from pancreatic resection of RCC despite the invasiveness of the surgery. Methods: Between January 1996 and December 2007, data from 744 patients were collected in a prospective pancreatic surgery database, and patients with metastasis into the pancreas from RCC were identified. Results: Resective surgery was performed in 14 patients with metastasis to the pancreas from RCC. Most patients were clinically asymptomatic. The median interval between primary treatment of RCC and occurrence of pancreatic metastasis was 94 months (range 32–158). The morbidity rate was 42.8%. Patients with a metastasis size <2.5 cm had a much better survival after resection (100 months) than those with a metastasis size >2.5 cm (44 months). Moreover, the number of métastases predicts the survival after resection. Conclusions: In patients with pancreatic métastases from RCC who have only limited disease, complete resection of all lesions can be successfully performed with a low rate of complications. Thus, patients with a history of RCC should be monitored for more than 10 years after nephrectomy to detect recurrence.
Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich.

Dvylikapirštę žarną išsauganti kasos rezekcija Frey būdu yra standartinė operacija gydant sergančiuosius komplikuotu lėtiniu pankreatitu. Kasos ir plonosios žarnos jungtis įprastai atliekama dviejų aukštų siūlėmis. Disertacinio darbo tikslas – palyginti Frey operacijos, naudojant vieno arba dviejų aukštų kasos – tuščiosios žarnos siūlę, rezultatatus, taip pat įvertinti operuotų pacientų gyvenimo kokybės pokyčius, praėjus 12 mėn. po operacijos. Perspektyviniame atsitiktinių imčių klinikiniame tyrime dalyvavo aštuoniasdešimt pacientų. Tiriamieji prieš operaciją atsitiktine tvarka suskirstyti į dvi grupes: pirmos grupės pacientams atlikta operacija, formuojant kasos – tuščiosios žarnos jungtį vieno aukšto ištisine siūle; antros grupės pacientams, kasos – tuščiosios žarnos jungtis suformuota dviejų aukštų pavienėmis siūlėmis. Tyrime nustatėme, kad bendras operacijos laikas (208±46 min ir 255±58 min), bei pankreojejunoanastomozės siuvimo laikas (19±6 min. ir 51±18 min.) buvo statistiškai reikšmingai mažesnis „vieno aušto siūlės“ grupėje nei „dviejų aukštų pavienių siūlių“ grupėje. Pooperacinės komplikacijos, kasos fistulės dažnis, pooperacinė hospitalizavimo trukmė abiejose grupėse nesiskyrė. Vertinant gyvenimo kokybės pokyčius, praėjus 12 mėn. po operacijos, nustatytas statistiškai reikšmingas gyvenimo kokybės pagerėjimas abiejose pacientų grupėse.
The duodenum-preserving pancreatic resection according to Frey is a standard operation for patients with complicated chronic pancreatitis. The pancreatojejunostomy is usually performed using two layer suture. The aim of doctoral dissertation was to compare single- and two-layer suture in pancreatojejunostomy performed in duodenum-preserving pancreatic resection according to Frey modification and to assess the changes of the quality of life 12 months after operation. A total of 80 patients were enrolled in the prospective randomized clinical. They were randomly allocated into two groups. In the first group of patients, pancreatojejunostomy was constructed by using single-layer continuous suture. In the second group of patients, pancreatojejunostomy was constructed by using two-layer interrupted suture. Overall time of the operation (208 ± 46 min. and 255 ± 58 min.) and the suturing time (19 ± 6 min. and 51 ± 18 min.) were significantly shorter in the single layer anastomosis group. Postoperative complications, the prevalence of pancreatic fistula, the length of stay did not differ in both groups. There was a statistically significant improvement of the quality of life 12 months after operation in the both groups of patients.

The duodenum-preserving pancreatic resection according to Frey is a standard operation for patients with complicated chronic pancreatitis. The pancreatojejunostomy is usually performed using two layer suture. The aim of doctoral dissertation was to compare single- and two-layer suture in pancreatojejunostomy performed in duodenum-preserving pancreatic resection according to Frey modification and to assess the changes of the quality of life 12 months after operation. A total of 80 patients were enrolled in the prospective randomized clinical. They were randomly allocated into two groups. In the first group of patients, pancreatojejunostomy was constructed by using single-layer continuous suture. In the second group of patients, pancreatojejunostomy was constructed by using two-layer interrupted suture. Overall time of the operation (208 ± 46 min. and 255 ± 58 min.) and the suturing time (19 ± 6 min. and 51 ± 18 min.) were significantly shorter in the single layer anastomosis group. Postoperative complications, the prevalence of pancreatic fistula, the length of stay did not differ in both groups. There was a statistically significant improvement of the quality of life 12 months after operation in the both groups of patients.
Dvylikapirštę žarną išsauganti kasos rezekcija Frey būdu yra standartinė operacija gydant sergančiuosius komplikuotu lėtiniu pankreatitu. Kasos ir plonosios žarnos jungtis įprastai atliekama dviejų aukštų siūlėmis. Disertacinio darbo tikslas – palyginti Frey operacijos, naudojant vieno arba dviejų aukštų kasos – tuščiosios žarnos siūlę, rezultatatus, taip pat įvertinti operuotų pacientų gyvenimo kokybės pokyčius, praėjus 12 mėn. po operacijos. Perspektyviniame atsitiktinių imčių klinikiniame tyrime dalyvavo aštuoniasdešimt pacientų. Tiriamieji prieš operaciją atsitiktine tvarka suskirstyti į dvi grupes: pirmos grupės pacientams atlikta operacija, formuojant kasos – tuščiosios žarnos jungtį vieno aukšto ištisine siūle; antros grupės pacientams, kasos – tuščiosios žarnos jungtis suformuota dviejų aukštų pavienėmis siūlėmis. Tyrime nustatėme, kad bendras operacijos laikas (208±46 min ir 255±58 min), bei pankreojejunoanastomozės siuvimo laikas (19±6 min. ir 51±18 min.) buvo statistiškai reikšmingai mažesnis „vieno aušto siūlės“ grupėje nei „dviejų aukštų pavienių siūlių“ grupėje. Pooperacinės komplikacijos, kasos fistulės dažnis, pooperacinė hospitalizavimo trukmė abiejose grupėse nesiskyrė. Vertinant gyvenimo kokybės pokyčius, praėjus 12 mėn. po operacijos, nustatytas statistiškai reikšmingas gyvenimo kokybės pagerėjimas abiejose pacientų grupėse.

Contemporary hepatobiliary surgery practice must accurately assess operative risk in increasingly elderly populations with greater co-morbidity. Current methods fail to identify patients at high risk of postoperative complications. Cardiopulmonary exercise testing (CPET) derived anaerobic threshold (AT) and ventilatory equivalence of carbon dioxide (VE/VCO2) are validated predictors of postoperative outcome in major intra-abdominal surgery and outperform contemporary tools of risk evaluation. Despite evidence of improved in-hospital postoperative survival in large centres offering complex curative hepatobiliary surgery, morbidity remains high and long-term survival in the high-risk subset remains poor. This thesis investigated the role of validated CPET-derived markers in predicting perioperative outcomes for a high-risk hepatobiliary surgery population. It was also utilised to study the impact of malignant obstructive jaundice on peripheral oxygen extraction. In a prospective cohort of high-risk patients undergoing liver resection, an AT of 9.9 ml O2/kg/min predicted in-hospital mortality and long-term survival. Below this threshold, AT was 100% sensitive and 75.9% specific for in-hospital mortality (PPV 19%, NPV 100%). Long-term survival below the threshold of 9.9 was significantly worse when compared to those above (mortality HR 1.81). The VE/VCO2 was the most significant predictor of postoperative complications and a threshold of 34.5 provided 84% specificity and 47% sensitivity (PPV 76%, NPV 60%). Amongst the high-risk pancreaticoduodenectomy patients, VE/VCO2 was the single most predictive marker of in-hospital postoperative mortality with an AUC of 0.850 (p=0.020); a threshold value 41 was 75% sensitive and 94.6% specific (PPV 50%, NPV 98.1%). The VE/VCO2 41 was also the only predictor of poor long-term survival (HR 1.90). Notably, AT, Revised Cardiac Risk Index and Glasgow Prognostic Score did not predict outcome after pancreaticoduodenectomy. Patients with malignant obstructive jaundice, evaluated for peripheral oxygen extraction using CPET, showed lower mean peak oxygen consumption (peak VO2) at 63±17.4% of the predicted value. This was noted in absence of any significant pre-existing cardiopulmonary disease and normal respiratory reserve. Normal patterns of oxygen extraction were seen at rest, during incremental work rate and peak exercise levels. Levels of oxygen partial pressure and saturation exceeded baseline values after exercise signifying normal microcirculatory responses. Thus, aerobic capacity was limited by dysfunction in delivery (cardiac output) rather than oxygen extraction. CPET provides useful prognostic adjuncts for early and long-term outcomes in the high-risk patients undergoing major hepatobiliary surgery. These findings provide useful tools for perioperative optimisation of the high-risk patient and plan appropriate level of postoperative care to address mortality and morbidity after surgery.

碩士
國立陽明大學
醫務管理研究所
99
Background: Some domestic and foreign academic researches indicate that although the changes of medical expenses regime have decreased the average medical expenses by making the patients discharged from hospital earlier, the subsequently increasing rate of re-hospitalization have enlarged the consumption of medical resources（Enrico Brizioli et al., 1996）. Goal : The goal was to understand the number of times of re-hospitalization, the duration of re-hospitalization and the reasons of re-hospitalization of the patients who had undergone pancreatectomy within one year and to analyze the factors related to such re-hospitalization in order to improve the care quality after pancreatectomy. Methods: This is a retrospective study. The data source was from the National Health Insurance academic research database from 2004 to 2005. This study focused on the patients who had undergone pancreatectomy and integrated and analyzed their medical files. The goal was to investigate the frequency and reasons of re-hospitalization of the patients through their hospitalization records within 14 days, within from 15 to 30 days, within from 31 to 90 days, within from 91 to 180 days, within from 181 days to one year, within 90 days, within from 90 days to one year, including the state of health, surgical factors, medical resources, to assess the relationship between different factors and re-hospitalization, quantify the risk of the above-mentioned factors with respect to the re-hospitalization and compare the reasons and risk factors of the re-hospitalization at different times. Furthermore, the re-hospitalization within 90 days of a patient was defined early re-hospitalization and the re-hospitalization within from 90 days to one year was defined late re-hospitalization; and, this study assessed the impact of early and late re-hospitalization on the risk factors. Discussions: This study was conducted on the basis of a total of 1,317 samples. With respect to the impact of the re-hospitalization at different times, this study used Logistic Regression to analyze the relevant factors to identify the risk factors. The analysis result showed that there were obvious differences with respect to gender, Charlson Co-morbidity Index, amount of physician services, level of the hospital which performed such surgery, complication caused by such surgery, and such disease identified as pancreatic cancer. As to the impact of the re-hospitalization within one year after the surgery, this study used Poisson Regression to analyze the relevant factors to identify the risk factors. The analysis result showed that there were obvious differences with respect to gender, Charlson Co-morbidity Index, number of days of such hospitalization, amount of physician services, level of the hospital which performed such surgery, complication caused by such surgery, and such disease identified as pancreatic cancer. As for the impact of the re-hospitalization at different times after the surgery and the early and late re-hospitalization due to various diseases, this study used Logistic Regression to analyze the relevant factors. Because there were only a small number of cases of re-hospitalization at different times due to surgery-related complications (please see Table 4-1), the relevant risk factors were not discussed here. Besides, because the numbers of cases of re-hospitalization at different times were different, all the relevant factors were put into the model to use stepwise regression method to select significant risk factors (α=0.2). The result of the statistical analysis will be described in detail in this study report. Conclusions: The study result indicates that: (1) There were less cases of re-hospitalization due to pancreatic disease within 90 days after the surgery (early re-hospitalization) when a patient was one year younger. (2) The result of care was better when there was a larger amount of physician services. (3) Where such disease was identified as pancreatic cancer, the number of cases of re-hospitalization within 30 days after the surgery of such disease was larger than the one that was not identified as pancreatic cancer. (4) Where a patient’s Charlson Co-morbidity Index before the surgery was larger, there were a larger number of cases of re-hospitalization within 90 days after the surgery (early re-hospitalization) and a larger number of cases of re-hospitalization due to metastatic disease of cancer. (5) Compared to the patients who had undergone pancreatic tail resection, the patients who had undergone Whipple Procedure (Whipple pancreaticoduodenectomy) had a larger number of cases of re-hospitalization within 90 days after the surgery (early re-hospitalization) due to metastatic disease of cancer but had a larger number of cases of re-hospitalization within from 91 days to one year after the surgery (late re-hospitalization) due to pancreatic cancer.

Background: Pancreatic mucinous cystic neoplasms (MCN) are rare mucin-producing cystic tumors. They are predominantly found, incidentally, in middle-aged women and usually located in the pancreatic body or tail. They are differentiated from other mucin producing neoplasms by the presence of ovarian-type stroma. The current management of MCN is defined by the consensus European, International Association of Pancreatology (IAP) and the American Association of Gastroenterology guidelines. However, the malignant potential of these lesions remains uncertain, with differing rates of malignant potential being described. Since the criteria for surgical resection differs between the current guidelines, the aims of this large multi-institution study were to determine the rate of associated malignancy in resected MCNs and to determine predictor features, clinical and radiological, for malignant transformation in MCN. Methods: All surgically resected MCNs between January 2003 and December 2015 were included in this international multicentre retrospective study. Lesions without ovarian type stroma were excluded. All lesions found in men had the diagnosis of MCN confirmed by two experienced pancreatic pathologists. Malignant MCNs were defined by the presence of invasive adenocarcinoma. Results: 211 patients with a confirmed and surgically resected MCN were included. Median age was 53 (range 18–82) years, and 95.7% (202/211) were in women. Median pre-operative tumour size was 52 (range 12-230) mm. 16.1% (34/211) were malignant. The rates of malignancy (33.3% (3/9) vs. 15.3% (31/202)) and high-grade dysplasia (33.3% (3/9) vs. 15.8% (32/202) were double in men compared to women. In all cases of malignancy or high-grade dysplasia, at least one of the following characteristics was seen: male patient, symptoms, or a preoperative worrisome feature (solid component, septations, main pancreatic duct dilatation >6mm, elevated serum ca 19-9). A total of five cases of malignant transformation occurred in MCNs less than 4 cm in size. All these cases were associated with features of concern on pre-operative cross-sectional imaging. Conclusion: In female patients in this large multicentre study, malignancy or high-grade dysplasia was solely seen in MCNs with symptoms or worrisome features on preoperative imaging, regardless of the size of the tumour. In males, the risk of malignancy was significantly higher than in females, suggesting that operative treatment should be considered in all male patients with a suspected MCN of any size. In female patients, conservative management seems to be a safe approach for suspected MCNs of any size without symptoms or worrisome features.

碩士
國立臺灣大學
流行病學研究所
97
ABSTRACT Purpose: Liver recurrence develops in 60% of patients who undergo resection for pancreatic cancer (PC) and predicts a dismal prognosis. Experimental evidences suggested chemokine receptor CXCR4 as the key mediator of liver metastasis in PC, but its significance has not been investigated with patient outcome. This study aimed to investigate the potential associations between CXCR4 expression and liver recurrence or overall survival after resection for PC. Methods: Ninety-seven patients undergoing R0 resection were evaluated. CXCR4 expression was analyzed by immunohistochemistry, and its association with liver recurrence-free or overall survival was analyzed by Kaplan-Meier estimates and multivariable proportional hazards models. Results: Patients with CXCR4-positive tumors had worse prognosis than those with CXCR4-negative tumors, with a shorter liver recurrence-free survival (median: 8.7 vs. 39.7 months; p=0.004) and overall survival (median: 10.2 vs. 22.3 months; p<0.001). Overall survival for CXCR4-positive stage IIa patients was similar to stage IIb patients and significantly shorter than CXCR4-negative stage IIa patients (p=0.002). The adjusted hazard ratio of positive CXCR4 immunostaining was 2.22 for liver recurrence (p=0.018), and 1.78 for death due to PC (p=0.041), respectively. Conclusion: CXCR4 expression is an independent predictor of early liver recurrence and death after resection for PC. CXCR4 immunohistochemistry provides exclusive prognostic information that can not be replaced by known prognostic factors and supplements TNM stage in predicting survival.

Le pseudocisti pancreatiche sono fra le complicanze più frequenti della pancreatite cronica, di quella acuta e dei traumi pancreatici. Da un punto di vista istopatologico, una pseudocisti pancreatica può essere descritta come una cavità a contenuto liquido delimitata da una parete costituita da tessuto fibroso o di granulazione, ma priva di un rivestimento epiteliale. Il meccanismo fisiopatologico che induce la formazione di una pseudocisti consiste nella rottura del sistema duttale pancreatico con fuoriuscita di succo pancreatico nel parenchima ghiandolare; la reazione infiammatoria che segue si organizza formando la parete fibrosa. Questa rottura può essere secondaria ad infiammazione e necrosi (pancreatite), o ad un aumento della pressione intraduttale (litiasi) o ad un trauma, che può essere anche secondario ad un intervento chirurgico. Le pseudocisti pancreatiche hanno una presentazione clinica molto variabile che va da lesioni completamente asintomatiche a formazioni che causano dolore addominale, sindrome subocclusiva da compressione gastrica o duodenale e possono presentarsi come una massa palpabile o, più raramente, con ittero. Uno dei principali fattori che influenza la gestione delle pseudocisti risiede nell’eziologia delle pseudocisti stesse. Quelle che si formano in seguito a pancreatite acuta possono risolversi spontaneamente in 4 o 6 settimane, con un completo riassorbimento della componente liquida e “restituito ad integrum” della ghiandola pancreatica e dei tessuti circostanti. Nella pancreatite cronica invece la risoluzione spontanea avviene raramente, in quanto la parete della pseudocisti ha il tempo di organizzarsi in modo definitivo con una componente fibrotica. In letteratura viene riportata una percentuale di risoluzione spontanea del 25% in corso di pancreatite cronica. Le indicazioni per una terapia interventistica riguardano le pseudocisti complicate, quelle sintomatiche oppure le asintomatiche con diametro > 5 cm, con dimensioni stabili da più di 6 settimane; è infatti altamente improbabile avere una risoluzione spontanea di una pseudocisti dopo 6 settimane dalla comparsa e maggiori sono le dimensioni, maggiori sono i rischi di complicanze gravi quali emorragia, infezione o rottura della cisti. Per quanto riguarda le opzioni terapeutiche, il trattamento chirurgico rappresenta storicamente la terapia di scelta. Ancora oggi, nonostante lo sviluppo di tecniche mini-invasive, la chirurgia ha un ruolo importante soprattutto nel trattamento di pseudocisti complicate da necrosi e infezione, e nelle lesioni cistiche sospette. L’approccio endoscopico alle pseudocisti era già stato tentato negli anni ’70, ma ha conosciuto una significativa espansione solo negli anni ’90. Il primo trattamento endoscopico fu eseguito nel 1973 quando Roger et al. praticarono l’aspirazione di una pseudocisti attraverso lo stomaco. Questa procedura non fu, però, risolutiva, in quanto il liquido si riaccumulava riformando la cisti. Con gli anni la tecnica è andata raffinandosi, fino all’attuale inserimento di una endoprotesi per drenare le raccolte fino al completo collabimento. La pseudocisti deve avere una parete matura (spessore > 3 mm e < 1 cm), deve essere in intimo contatto con la parete dello stomaco e/o duodeno determinandovi anche un’impronta riconoscibile e avere un diametro di almeno 5 o 6 cm. Sarebbe opportuno, inoltre, valutare la presenza di pseudoaneurismi, di ipertensione portale e circoli collaterali per ridurre il rischio di sanguinamento dopo l’incisione della parete. Esistono tre approcci endoscopici al trattamento delle pseudocisti pancreatiche: 1) APPROCCIO TRANSMURALE: consiste nel drenaggio della raccolta posizionando una protesi attraverso la parete gastrica o duodenale; è possibile quando la cisti determina un’impronta ben visibile sulla parete. 2) APPROCCIO TRANSPAPILLARE: consiste nel posizionamento di uno stent attraverso la papilla di Vater avanzando fino a superare, e in tal modo bypassare, la lesione del dotto. Questo approccio viene scelto per drenare le pseudocisti che comunicano con il dotto pancreatico principale e che non determinano un’impronta sullo stomaco o sul duodeno. 3) APPROCCIO TRANSMURALE EUS-GUIDATO: consiste nello svuotamento della raccolta tramite uno stent posizionato nella parete intestinale, sotto il controllo di uno strumento ecoendoscopico. Quest’ultimo consente di individuare cisti < 2 cm, offrendo quindi una maggior sicurezza anche in pazienti che non presentano né l’impronta gastrica né una comunicazione con i dotti. All’ecoendoscopia può essere associato uno studio Doppler dei vasi della parete gastrica, che permette di valutare la presenza di malformazioni dei vasi (varici e aneurismi) riducendo il rischio di sanguinamento. Nonostante l’approccio endoscopico sia in uso dagli anni 80, non esistono ancora linee guida ed un consenso unanime nel definire il suo ruolo nel trattamento delle pseudocisti. Gli studi pubblicati mostrano un alta percentuale di successo con bassi tassi di morbidità, di mortalità e di ricorrenza e migliori risultati rispetto a quelli riportati dopo chirurgia. Tuttavia, questi sono tutti studi retrospettivi e principalmente analizzano singole esperienze con casistiche piccole e con differenze enormi fra tecniche di drenaggio utilizzate. La maggior parte di queste serie sono inoltre eterogenee, includendo pazienti affetti da pseudocisti a diversa eziopatogenesi. Soltanto pochi studi riportano risultati in gruppi omogenei di pazienti, quali pazienti con pancreatite acuta o cronica. Questi studi hanno evidenziato che i pazienti trattati per pseudocisti insorte a seguito di pancreatite cronica hanno risultati migliori rispetto a quelli colpiti da pancreatite acuta. Finora nessuno studio ha analizzato l’approccio endoscopico nelle pseudocisti insorte a seguito di resezioni pancreatiche per neoplasia. Lo scopo del nostro studio è di contribuire alla valutazione della sicurezza e dell'efficacia del trattamento endoscopico delle pseudocisti in questa popolazione specifica di pazienti. Inoltre abbiamo paragonato il risultato ottenuto con quello dei pazienti sottoposti nello stesso periodo al medesimo trattamento, ma per pseudocisti insorte a seguito di una malattia infiammatoria del pancreas. Da gennaio 1999 a giugno 2008 sono stati inclusi in questo studio 55 pazienti. Tutti i dati clinici sono stati esaminati retrospettivamente. In 25 Pazienti la pseudocisti era la complicanza insorta a seguito di intervento chirurgico per asportazione di una neoplasia del pancreas. In 30 pazienti la pseudocisti era insorta a seguito di patologia infiammatoria (pancreatite acuta severa in 28 pazienti e pancreatite cronica in 2). Durante il follow-up (mediana: 34 mesi) la percentuale di successo del trattamento endoscopico è stata leggermente superiore nel gruppo chirurgico (84%) rispetto a quello infiammatorio (73%), anche se non in maniera statisticamente significativa (p = 0.532). In 4 pazienti del gruppo chirurgico la procedura non è stata efficace. Un Paziente ha necessitato di un drenaggio percutaneo, uno di laparotomia per rimuovere una pseudocisti infetta, uno di laparotomia per emorragia gastrica ed un altro di un ulteriore drenaggio endoscopico. La procedura è stata inefficace in 8 casi del gruppo post-infiammatorio; in 6 casi è stato necessario un intervento chirurgico mentre un caso non è stato comunque più trattato perché asintomatico nonostante la pseudo cisti permanesse. Un paziente infine è deceduto per emorragia massiva da rottura di uno pseudoaneurisma dell'arteria gastroduodenale. In altri 6 pazienti si è verificata una ricorrenza della pseudocisti ed è stato necessario un secondo trattamento endoscopico, con successo definitivo in cinque casi. In conclusione il trattamento endoscopico delle pseudocisti che insorgono a seguito di una resezione per neoplasia pancreatica è una tecnica affidabile ed efficace, associata a basso tasso di complicanze e alta percentuale di successo.
BACKGROUND: Endoscopy has been regarded as an effective modality for draining pancreatic collections, pseudocysts, and abscesses. This study analyzes our experience with endoscopic transmural drainage of pancreatic pseudocysts and compares the outcomes in patients with postsurgical and pancreatitis-associated ones. METHODS: Patients who underwent endoscopic drainage of a pancreatic pseudocyst from January 1999 through June 2008 were included in this retrospective analysis. The specific indication for attempting the procedure was the presence of direct contact between the pseudocyst and the gastric wall. All the drainages were carried out via a transgastric approach, and one or two straight plastic stents (10 or 11.5 French) were positioned. A comparative analysis of short- and long-term results was made between patients with postoperative pseudocysts (group A) and patients with pancreatitis-associated pseudocysts (group B). RESULTS: Fifty-five patients were included in the study, 25 in group A and 30 in group B. Overall, a single stent was inserted in 84.0% of patients, while two stents were needed in the remaining 16.0%. The technical success rate was 78.2%, whereas procedure-related complications were 16.4%. Complications included pseudocyst superinfection and major bleeding and were managed mainly by surgery. Mortality rate was 1.8% (1 patient). There were no significant differences in the technical success rate and procedure-related complications between the two groups (p = 0.532 and 0.159, respectively) Recurrences were 13.9% and significantly more common in group B (p = 0.021). In such cases, a second endoscopic drainage was successfully performed. CONCLUSION: Transmural endoscopic treatment of pancreatic pseudocysts is feasible and has a technical success rate of 78.2%, without differences related to the pseudocyst etiology. Recurrences, on the other hand, are more common in patients with pancreatitis. Given the severe complications that may occur after the procedure, we recommend that endoscopic drainage be performed in a tertiary-care center with specific expertise in pancreatic surgery.

RESUMO: Actualmente, a única possibilidade de cura para doentes com adenocarcinoma do pâncreas (PDAC) é a ressecção cirúrgica, no início deste estudo, perguntamo-nos se os predictores clínico-patológicos clássicos de prognostico poderiam ser validados em uma grande cohort de doentes com cancro do pâncreas ressecável e se outros predictores clínicos poderiam ter um papel na decisão de que doentes beneficiariam de ressecção cirúrgica. No capítulo 2, observamos que até 30% dos doentes morrem no primeiro ano após a ressecção cirúrgica, pelo que o nosso objectivo foi determinar factores pré-operatórios que se correlacionam com mortalidade precoce após ressecação cirúrgica com recurso a um instrumento estatisticamente validado, o Charlson-Age Comorbidity Index (CACI), determinamos que um CACI score superior a 4 foi preditivo de internamentos prolongados (p <0,001), complicações pós-operatórias (p = 0,042), e mortalidade em 1 ano pós- ressecção cirúrgica (p <0,001). Um CACI superior a 6 triplicou a mortalidade no primeiro ano pós-cirurgia e estes doentes têm menos de 50% de probabilidade de estarem vivos um ano após a cirurgia. No capítulo 3, o nosso objectivo foi identificar uma proteína de superfície que se correlacionasse estatisticamente com o prognostico de doentes com adenocarcinoma do pâncreas e permitisse a distinção de subgrupos de doentes de acordo com as suas diferenças moleculares, perguntamo-nos ainda se essa proteína poderia ser um marcador de células-estaminais. No nosso trabalho anterior observamos que as células tumorais na circulação sanguínea apresentavam genes com características bifenotípica epitelial e mesenquimal, enriquecimento para genes de células estaminais (ALDH1A1 / ALDH1A2 e KLF4), e uma super-expressão de genes da matriz extracelular (colagénios, SPARC, e DCN) normalmente identificados no estroma de PDAC. Após a avaliação dos tumores primários com RNA-ISH, muitos dos genes identificados, foram encontrados co-localizando em uma sub-população de células na região basal dos ductos pancreáticos malignos. Além disso, observamos que estas células expressam o marcador SV2A neuroendócrino, e o marcador de células estaminais ALDH1A1/2. Em comparação com tumores negativos para SV2, os doentes com tumores SV2 positivos apresentaram níveis mais baixos de CA 19-9 (69% vs. 52%, p = 0,012), tumores maiores (> 4 cm, 23% vs. 10%, p = 0,0430), menor invasão de gânglios linfáticos (69% vs. 86%, p = 0,005) e tumores mais diferenciados (69% vs. 57%, p = 0,047). A presença de SV2A foi associada com uma sobrevida livre de doença mais longa (HR: 0,49 p = 0,009) bem como melhor sobrevida global (HR: 0,54 p = 0,018). Em conjunto, esta informação aponta para dois subtipos diferentes de adenocarcinoma do pâncreas, e estes subtipos co-relacionam estatisticamente com o prognostico de doentes, sendo este subgrupo definido pela presença do clone celular SV2A / ALDH1A1/2 positivo com características neuroendócrinas. No Capítulo 4, a expressão de SV2A no cancro do pâncreas foi validado em linhas celulares primárias. Demonstramos a heterogeneidade do adenocarcinoma do pâncreas de acordo com características clonais neuroendócrinas. Ao comparar as linhas celulares expressando SV2 com linhas celulares negativas, verificamos que as linhas celulares SV2+ eram mais diferenciadas, diferindo de linhas celulares SV2 negativas no que respeita a mutação KRAS, proliferação e a resposta à quimioterapia. No capítulo 5, perguntamo-nos se o clone celular SV2 positivo poderia explicar a resistência a quimioterapia observada em doentes. Observamos um aumento absoluto de clones celulares expressando SV2A, em múltiplas linhas de evidência - doentes, linhas de células primárias e xenotransplantes. Embora, tenhamos sido capazes de demonstrar que o adenocarcinoma do pâncreas é uma doença heterogénea, consideramos que a caracterização genética destes clones celulares expressando SV2A é de elevada importância. Pretendemos colmatar esta limitação com as seguintes estratégias: Após o tratamento com quimioterapia neoadjuvante na nossa coorte, realizamos microdissecação a laser das amostras primarias em parafina, de forma a analisar mutações genéticas observadas no adenocarcinoma pancreático; em segundo lugar, pretendemos determinar consequências de knockdown da expressão de SV2A em nossas linhas celulares seguindo-se o tratamento com gemicitabina para determinação do papel funcional de SV2A; finalmente, uma vez que os nossos esforços anteriores com um promotor - repórter e SmartFlare ™ falharam, o próximo passo será realizar RNA-ISH PrimeFlow™ seguido de FACS e RNA-seq para caracterização deste clone celular. Em conjunto, conseguimos provar com várias linhas de evidência, que o adenocarcinoma pancreático é uma doença heterogénea, definido por um clone de células que expressam SV2A, com características neuroendócrinas. A presença deste clone no tecido de doentes correlaciona-se estatisticamente com o prognostico da doença, incluindo sobrevida livre de doença e sobrevida global. Juntamente com padrões de proliferação e co-expressão de ALDH1A1/2, este clone parece apresentar um comportamento de células estaminais e está associado a resistência a quimioterapia, uma vez que a sua expressão aumenta após agressão química, quer em doentes, quer em linhas de células primárias.----------------------------- ABSTRACT: Currently, the only chance of cure for patients with pancreatic adenocarcinoma is surgical resection, at the beginning of my thesis studies, we asked if the classical clinicopathologic predictors of outcome could be validated in a large cohort of patients with early stage pancreatic cancer and if other clinical predictors could have a role on deciding which patients would benefit from surgery. In chapter 2, we found that up to 30% of patients die within the first year after curative intent surgery for pancreatic adenocarcinoma. We aimed at determining pre-operative factors that would correlate with early mortality following resection for pancreatic cancer using a statistically validated tool, the Charlson-Age Comorbidity Index (CACI). We found that a CACI score greater than 4 was predictive of increased length of stay (p<0.001), post-operative complications (p=0.042), and mortality within 1-year of pancreatic resection (p<0.001). A CACI score of 6 or greater increased 3-fold the odds of death within the first year. Patients with a high CACI score have less than 50% likelihood of being alive 1 year after surgery. In chapter 3 we aimed at identifying a surface protein that correlates with patient’s outcome and distinguishes sub-groups of patients according to their molecular differences and if this protein could be a cancer stem cell marker. The most abundant class of circulating tumor cells identified in our previous work was found to have biphenotypic features of epithelial to mesenchymal transition, enrichment for stem-cell associated genes (ALDH1A1/ALDH1A2 and KLF4), and an overexpression of extracellular matrix genes (Collagens, SPARC, and DCN) normally found in the stromal microenvironment of PDAC primary tumors. Upon evaluation of matched primary tumors with RNA-ISH, many of the genes identified were found to co-localize in a sub-population of cells at the basal region of malignant pancreatic ducts. In addition, these cells expressed the neuroendocrine marker SV2A, and the stem cell marker ALDH1A1/2. Compared to SV2 negative tumors, patients with SV2 positive tumors were more likely to present with lower CA 19-9 (69% vs. 52%, p = 0.012), bigger tumors (size > 4 cm, 23% vs. 10%, p= 0.0430), less nodal involvement (69% vs. 86%, p = 0.005) and lower histologic grade (69% vs. 57%, p = 0.047). The presence of SV2A expressing cells was associated with an improved disease free survival (HR: 0.49 p=0.009) and overall survival (HR: 0.54 p=0.018) and correlated linearly with ALDH1A2. Together, this information points to two different sub-types of pancreatic adenocarcinoma, and these sub-types correlated with patients’ outcome and were defined by the presence of a SV2A/ ALDH1A1/2 expressing clone with neuroendocrine features. In Chapter 4, SV2A expression in cancer was validated in primary cell lines. We were able to demonstrate pancreatic adenocarcinoma heterogeneity according to neuroendocrine clonal features. When comparing SV2 expressing cell lines with SV2 negative cell lines, we found that SV2+ cell lines were more differentiated and differ from SV2 negative cell lines regarding KRAS mutation, proliferation and response to chemotherapy. In Chapter 5 we aimed at determining if this SV2 positive clone could explain chemoresistance observed in patients. We found an absolute increase in SV2A expressing cells, with multiple lines of evidence, in patients, primary cell lines and xenografts. Although, we have been able to show evidence that pancreatic adenocarcinoma is a heterogeneous disease, our findings warrant further investigation. To further characterize SV2A expressing clones after treatment with neoadjuvant chemotherapy in our cohort, we have performed laser capture microdissection of the paraffin embedded tissue in this study and will analyze the tissue for known genetic mutations in pancreatic adenocarcinoma; secondly, we want to know what will happen after knocking down SV2A expression in our cell lines followed by treatment with gemcitabine to determine if SV2A is functionally important; finally, since our previous efforts with a promoter – reporter and SmartFlare™ have failed, we will utilize a novel PrimeFlow™ RNA-ISH assay followed by FACS and RNA sequencing to further characterize this cellular clone. Overall our data proves, with multiple lines of evidence, that pancreatic adenocarcinoma is a heterogeneous disease, defined by a clone of SV2A expressing cells, with neuroendocrine features. The presence of this clone in patients’ tissue correlates with patient’s disease free survival and overall survival. Together with patterns of proliferation and ALDH1A1/2 co-expression, this clone seems to present a stem-cell-like behavior and is associated with chemoresistance, since it increases after chemotherapy, both in patients and primary cell lines.

Nonostante la disponibilità di prove scientifiche per il fast- track concetto di chirurgia pancreatica sua traduzione nella pratica clinica, dalla maggior parte delle istituzioni, rimane lento. Motivi per essere la mancanza di consapevolezza dei dati accelerata basata sull'evidenza; una mancanza di accordo con i dati (difficili da accettare); mancanza di convinzione che la propria istituzione può realmente eseguire un intervento chirurgico fast-track, limiti di tempo e competenze insufficienti o personale di supporto. Tuttavia recenti scoperte all'interno di specifici componenti di assistenza perioperatoria del pancreas gestione postoperatoria potrebbe contribuire a convalidare ulteriormente chirurgia pancreatica fast- track e migliorare il suo adattamento. Metodi: Tra il gennaio 2011 e agosto 2013, i pazienti sottoposti a resezione pancreatica sono stati arruolati nello studio ad una delle due istituzioni. Il braccio Università di Verona sottoposto ad innalzare il protocollo post-operatorio di recupero Verona, mentre il National Cancer Institute, gruppo di Università del Cairo sono stati sottoposti ai tradizionali gestione postoperatoria. Entrambi i gruppi sono stati seguiti per un efficace controllo del dolore, di ricominciare precoce di alimentazione orale, efficace mobilitazione immediata e il ripristino della funzione intestinale dopo intervento chirurgico. Le misure di esito per ciascun gruppo di pazienti sono stati valutati in termini di complicanze postoperatorie, quali fistola pancreatica (PF), ritardato svuotamento gastrico, perdita biliare, ascesso intra-addominale, post-pancreatectomy emorragia, pancreatite acuta, infezione della ferita, mortalità a 30 giorni, postoperatoria tariffe per l'ospedale, e di riammissione. Risultati: morbilità generale per Verona (n = 101) e Il Cairo (n= 98) è stato del 35% e 44,6 %, rispettivamente; e la mortalità a 30 giorni è stata del 5,9 % contro 8,2%. In entrambi i gruppi postoperatoria PF era la complicazione associata più frequente. Abbiamo osservato 10 fistole nel gruppo di Verona ( 9,9 % ), e 32 in gruppo Cairo ( 32,7 % ) . Ritardato svuotamento gastrico si è verificato nel 5% dei pazienti Verona e il 10,2 % del Cairo. Tasso di riammissione è stata del 4 % ( Verona) e il 2,8 % (Il Cairo ) . La lunghezza complessiva del soggiorno, tenendo in considerazione le riammissioni, è rimasto significativamente più breve nel gruppo fast track ( mediana 9 giorni, range: 7-16 giorni contro 14 giorni, range: 8-29 giorni ; p < 0.001 ). La destinazione di scarico primario era a casa in entrambi i gruppi. Conclusioni: Le prove disponibili e dati rispetto ai risultati, fornire una serie di raccomandazioni per suggerire alcuni elementi per un protocollo standardizzato. I dati sulla lunghezza del soggiorno per entrambi i percorsi sono incoraggianti verso l'attuazione di un percorso standardizzato di gestione postoperatoria.
Background: Despite the availability of the scientific evidence for the pancreatic fast-track surgery concept its translation into clinical practice, by most institutions, remains slow. Reasons being lack of awareness of evidence-based fast-track data; a lack of agreement with the data (difficult to accept); lack of belief that their own institution can actually perform fast-track surgery, time-limitation and insufficient expertise or staff support. However recent findings within specific perioperative care components of pancreatic postoperative management could help further validate pancreatic fast-track surgery and enhance its adaptation. Methods: Between January 2011 and August 2013, patients who underwent pancreatic resection were enrolled into the study at either of the two institutions. The Verona University arm subjected to the Verona enhanced recovery postoperative protocol while the National Cancer Institute, Cairo University group were subjected to conventional postoperative management. Both groups were followed up for effective control of pain, early reinstitution of oral feeding, effective immediate mobilization and restoration of bowel function following surgery. Outcome measures for each patient group were assessed in terms of postoperative complications such as pancreatic fistula (PF), delayed gastric emptying, biliary leak, intra-abdominal abscess, post-pancreatectomy hemorrhage, acute pancreatitis, wound infection, 30-day mortality, postoperative hospital stay, and readmission rates. Results: Overall morbidity for Verona (n= 101) and Cairo (n= 98) was 35% and 44.6%, respectively; and 30-day mortality was 5.9% versus 8.2%. In both groups postoperative PF was the most frequent associated complication. We observed 10 fistulae in the Verona group (9.9%), and 32 in Cairo group (32.7%). Delayed gastric emptying occurred in 5% of Verona patients and 10.2% of Cairo. Readmission rate was 4% (Verona) and 2.8% (Cairo). The overall length of stay, taking into consideration readmissions, remained significantly shorter in the fast track group (median 9 days, range: 7-16 days versus 14 days, range: 8-29 days; p<0.001). The primary discharge destination was home in both groups. Conclusions: The available evidence and data when compared to the results, provide a set of recommendations to suggest some items for a standardized protocol. Data on length of stay for both pathways are encouraging towards implementing a standardized postoperative management pathway.

Lo scopo di questa tesi è stato di valutare la fattibilità, la sicurezza ed i risultati di Enhanced Recovery Protocol (ERP) per la pancreaticoduodenectomia e la pancreatectomia distale laparoscopica in un ospedale universitario di riferimento in Regno Unito. Specificamente per la pancreatectomia distale laparoscopica, l'obiettivo era anche quello di analizzare l'impatto della chirurgia laparoscopica e di ERP sui costi. Nella Parte I, Capitolo 2, abbiamo valutato la fattibilità e la sicurezza di ERP per la pancreaticoduodenectomia, in assenza di simili programmi pubblicati nel Regno Unito. La parte II e' incentrata sulla pancreatectomia distale. Nel capitolo 3 abbiamo valutato l'impatto dell'introduzione dell' approccio laparoscopico per la pancreatectomia distale e il suo impatto sui risultati e costi. Nel capitolo 4 abbiamo valutato se l'attuazione di uno specifico ERP per la pancreatectomia distale laparoscopica avrebbe potuto migliorare ulteriormente i risultati e costi. Nella Parte III, capitolo 5 di questa tesi, si sintetizzano i risultati principali, si illustra qual e' lo stato dell'arte e si discutono prospettive future. Nella parte IV i protocolli di ERP attualmente adottati presso University Hospital di Southampton per la pancreaticoduodenectomia e la pancreatectomia distale laparoscopica sono illustrati.
The aim of this thesis was to assess the feasibility, safety and outcomes of ERP for pancreaticoduodenectomy and laparoscopic distal pancreatectomy in a tertiary referral UK university hospital. Specifically for laparoscopic distal pancreatectomy, the aim was also to analyze the impact of laparoscopic surgery and ERP on the cost economics. In Part I, Chapter 2, we evaluated the feasibility and safety of ERP for pancreaticoduodenectomy, at a time when no other evidence was available from the UK. Part II focuses on distal pancreatectomy. In Chapter 3 we assessed the impact of the introduction of the laparoscopic approach for distal pancreatectomy and its impact on outcomes and costs. In Chapter 4 we evaluated whether the implementation of a specific ERP for laparoscopic distal pancreatectomy could have improved further outcomes and costs. Part III, Chapter 5 of this thesis summarises the main finding, discusses where we stand and addresses future prospective. In Part IV the ERPs currently adopted at University Hospital Southampton for pancreaticoduodenectomy and laparoscopic distal pancreatectomy are illustrated.