Objective:To investigate the application value of TRIANGLE operation in radical resection of pancreatic cancer.Methods:The retrospective and descriptive study was conducted. The clinicopathological data of 30 patients with pancreatic cancer who underwent TRIANGLE operation in the First Affiliated Hospital of Nanjing Medical University from March 2020 to July 2020 were collected. There were 18 females and 12 males, aged from 41 to 79 years, with a median age of 65 years. After assessment of the resectability and dissection of the hepatoduodenal ligament, the superior mesenteric artery, celiac axis, common hepatic artery, portal vein and superior mesenteric vein of patients were exposed. According to the location of tumor, patients were performed pancreaticoduodenectomy, distal pancreatectomy or total pancreatectomy combined with dissection of putatively tumor-infiltrated lymphatic and neural tissue from the triangular space. During the TRIANGLE operation, it required to at least finish the skeletonization of right semi-circumference of the superior mesenteric artery and celiac axis in cases of pancreatico-duodenectomy, whereas the left semi-circumference in cases of distal pancreatectomy. In principle, both of the superior mesenteric artery and celiac axis were circumferentially skeletonized for cases of total pancreatectomy. Lymphatic, neural and fibrous tissues between the superior mesenteric artery and the celiac artery were completely dissected in all patients. Observation indicators: (1) surgical conditions and postoperative histopathological examination; (2) postoperative recovery situations; (3) follow-up. Regular follow-up was conducted by telephone interview or outpatient examinations to understand tumor recurrence and metastasis of patients up to January 2021. Measurement data with skewed distribution were represented as M (range). Count data were described as absolute numbers or percentages, Fisher exact probability method was used for comparison between groups. Results:(1) Surgical conditions and postoperative histopathological examination: all patients underwent open operation, including 21 cases of pancreaticoduodenec-tomy, 6 cases of distal pancreatectomy, 2 cases of total pancreatectomy, 1 case of mid-segment preserving pancreatectomy. There were 16 cases combined with portal vein-superior mesenteric vein resection and 3 cases combined with left adrenal resection. For the dissection area of superior mesenteric artery-celiac axis, one cycle was conducted in 5 of 30 patients, one and a quarter cycle in 8 patients, one and a half cycle in 8 patients, one and three quarters?two cycles in 9 patients. There were 16 cases with left gastric vein dissection and 14 cases with left gastric vein preservation. The operative time of 30 patients was 287 minutes (range, 165?495 minutes) and the volume of intraoperative blood loss was 275 mL(range, 50?800 mL). Nine patients received intraoperative transfusion of red blood cells or frozen plasma. Postoperative histopathological examination showed that the tumor diameter was 3.4 cm (range, 1.2?7.3 cm), the number of harvested lymph nodes was 20 (range, 9?35), the number of positive lymph nodes was 2 (0?19). Of the 30 patients,20 cases had moderately differentiated tumor and 10 cases had poorly differentiated tumor. R 0 resection was achieved in 9 patients, 1 mm R 1 resection in 17 patients, and R 1 resection in 4 patients. Postoperative pathological T stages: 3 of 30 patents were in stage T1, 18 cases were in stage T2, 5 cases were in stage T3, and 4 cases were in stage T4. Postoperative pathological N stages: 9 of 30 patents were in stage N0, 13 cases were in stage N1, and 8 cases were in stage N2. Postoperative pathological TNM stages: 2 of 30 patents were in stage Ⅰa, 2 cases were in stage Ⅰb, 3 case were in stage Ⅱa, 11 cases were in stage Ⅱb, 12 cases were in stage Ⅲ. (2) Postoperative recovery situations: 20 of 30 patients had postoperative complications, including 6 cases of Clavien-Dindo grade I complications, 9 cases of grade Ⅱ complications, 3 cases of grade Ⅲa complications, 2 cases of grade V complications. The incidence of postoperative diarrhea was 0 in cases with one cycle dissection of tissues around superior mesenteric artery-celiac axis, 1/8 in cases with one and a quarter cycle dissection, 4/8 in cases with one and a half cycle dissection, 9/9 in cases in with one and three quarters-two cycles dissection, showing a significant difference between the four groups ( P<0.05). The incidence of delayed gastric emptying was 5/16 in patients with left gastric vein dissection, versus 1/14 in patients with left gastric vein preservation, showing no significant difference between the two groups ( P>0.05). Nineteen of 30 patients received adjuvant chemotherapy.Twenty-eight of 30 patients were successfully discharged from hospital, with the duration of postoperative hospital stay of 15 days(range, 8?68 days) and the rest of 2 patients died. Three patients had unplanned readmission within postoperative 90 days. (3) Follow-up: all the 28 patients discharged from hospital were followed up for 1.0?9.0 months, with the median follow-up time of 6.5 months. During the follow-up, one patient with locally advanced pancreatic cancer had local recurrence, 9 patients had liver metastasis (including 4 cases with resectable pancreatic cancer, 4 cases with borderline resectable pancreatic cancer, 1 case with borderline resectable pancreatic cancer), and 1 patient with borderline resectable pancreatic cancer had peritoneal metastasis. The rest of 17 patients survived without tumor recurrence. Conclusions:TRIANGLE operation in the radical resection of pancreatic cancer has high radicality and low postoperative local recurrence, but a slightly high morbidity and mortality rate. Its long-term efficacy needs to be further evaluated. It is recommend that this procedure should be performed for selected patients after neoadjuvant chemotherapy in high-volume pancreatic surgery centers.

Radical resection remains as the key treatment to improve the prognosis of pancreatic cancer, and arterial invasion hinders radical resection for locally advanced lesions. More accurate assessment of resectability and selection of appropriate surgical techniques depend on better understanding of the anatomy, histology, and tumor biology of pancreatic cancer invasion to the artery. Traditional radiology assessment for artery involvement, based on the circumference of involved arteries, is not equivalent to pathological artery invasion. The depth of arterial invasion is more important than the circumference in evaluating resectability. Results of morphological observa-tion on arterial structure shows that both the feasibility of sub-adventitial divestment technique (SDT) and the necessity of arterial resection depend on whether the external elastic lamina of artery is invaded. The SDT can provide an opportunity for radical resection for pancreatic cancer with arterial invasion, while avoiding the high mortality and incidence of complications caused by arterial resection. A more precise assessment of tumor invasion depth and biological behavior will provide a more reliable basis for surgical decision for the treatment of locally advanced pancreatic cancer with arterial invasion.

Objective:To evaluate the value of left-sided uncinate process first approach in pancreaticoduodenectomy.Methods:The clinical data of 152 patients who underwent the left-sided uncinate process first approach during pancreaticoduodenectomy at Pancreas Center, the First Affiliated Hospital of Nanjing Medical University from January 2020 to December 2020 were analyzed retrospectively. There were 64 females and 88 males,with age( M( Q R)) of 62.0(14.7)years(range:16.0 to 84.0 years). The clinical date of 117 patients who underwent pancreaticoduodenectomy without using left-sided uncinate process first approach in the same period was selected as the control group,including 65 females and 52 males,with age of 64.0(13.0) years(range:13.0 to 84.0 years). Fisher exact probability method and t test were used to compare the data between the two groups,rank sum test was used for comparison of continuous variables between the two groups. Results:Pancreaticoduodenectomy was successfully performed in 152 patients in left-sided uncinate process first approach group. The operation time was 222.5(77.0) minutes(range:117.0 to 480.0 minutes),the time of uncinate process resection from left-side(the time from jejunum dissection to complete dissociation of the uncinate process) was 11.0(4.5) minutes(range:7.5 to 20.0 minutes),the time of pancreatic head resection (the time from jejunum dissection to pancreaticoduodenal specimen removal) was 26.0(8.5) minutes(range:20.0 to 41.0 minutes),the intraoperative blood loss was 200(150) ml(range:50 to 800 ml),and the intraoperative blood transfusion rate was 9.2% (14/152). Postoperative conditions:The postoperative hospital stay was 12 (9) d(range:6 to 55 d),the overall incidence of postoperative complications was 59.9%(91/152),and there was no perioperative death. Pathological results:The R0 resection rate of periampullary malignant tumor was 64.3%(77/112),with negative rate of uncinate process margin was 91.1%(102/112). The R0 resection rate of pancreatic ductal adenocarcinoma was 46.9%,with negative rate of uncinate process margin was 89.1%(57/64). Compared with the non-left-sided uncinate process first approach group(222.5(77.0) minutes, 9.2%(14/152)),the left-sided uncinate process first approach group had shorter operation time(246.0(94.0) minutes) ( Z=3.964, P<0.01),less intraoperative blood loss (18.8%(22/117))( Z=4.843, P<0.01),and lower intraoperative blood transfusion rate(χ2=5.248, P=0.029). However,there were no significant differences between two groups in postoperative hospital stay( Z=1.682, P=0.093),postoperative overall complications( P=0.549),R0 resection rate of periampullary malignant tumor(χ2=2.012, P=0.156),and negative rate of uncinate process margin(χ2=2.108, P=0.147). Conclusions:The "left-sided uncinate process first approach" could completely resect uncinate process under a direct vision,especially when the uncinate process was behind the superior mesenteric artery or beyond the left lateral margin of the superior mesenteric artery. The "left-sided uncinate process first approach" might increase the negative rate of uncinate process margin and R0 resection rate for periampullary malignant tumor.

Objective:To investigate the clinical efficacy of distal pancreatectomy with celiac axis resection(DP-CAR).Methods:A total of 89 consecutive patients (50 males and 39 females) who were diagnosed with pancreatic body cancer and underwent DP-CAR in Pancreas Center,First Affiliated Hospital of Nanjing Medical University between September 2013 and June 2022 were retrospectively reviewed. There were 50 males and 39 females,with age( M(IQR)) of 63(12) years(range:43 to 81 years). Perioperative parameters,pathology results and follow-up data of these patients were analyzed, χ2 or Fisher′s test for categorical data while the Wilcoxon test for quantitative data. Survival results were estimated by the Kaplan-Meier survival method. Results:Among 89 cases,cases combined with portal vein-superior mesenteric vein or organ resection accounted for 22.5% (20/89) and 42.7% (38/89),respectively. The operative time,blood loss and postoperative hospital stay were 270 (110) minutes,300 (300) ml and 13 (10) days,respectively. The overall morbidity rate was 67.4% (60/89) while the major morbidity was 11.2% (10/89). The increase rate in transient liver enzymes was 42.7% (38/89),3.4% (3/89) for liver failure,53.9% (48/89) for clinically relevant postoperative pancreatic fistula,1.1% (1/89) for bile leak,3.4% (3/89) for chylous leak of grade B and C,11.2% (10/89) for abdominal infection,9.0% (8/89) for postoperative hemorrhage of grade B and C,4.5% (4/89) for delayed gastric emptying,6.7% (6/89) for deep vein thrombosis,3.4% (3/89) for reoperation,4.5% (4/89)for hospital mortality,7.9% (7/89) for 90-day mortality. The pathological type was pancreatic cancer for all 89 cases and pancreatic ductal adenocarcinoma made up 92.1% (82/89). The tumor size was 4.8(2.0) cm, ranging from 1.5 to 12.0 cm. The number of lymph nodes harvested was 14 (13)(range:2 to 33),with a positive lymph node rate of 13.0% (24.0%). The resection R0 rate was 30.0% (24/80) and the R1 (<1 mm) rate was 58.8% (47/80). The median overall survival time was 21.3 months (95% CI: 15.6 to 24.3) and the median disease-free survival time was 19.1 months (95% CI: 11.7 to 25.1). The overall survival at 1-year and 2-year were 69.60% and 39.52%. The median survival time of 58 patients with adjuvant chemotherapy was 24.3 months (95% CI: 17.8 to 32.3) while that of 13 patients without any kind of adjuvant therapy was 8.4 months (95% CI: 7.3 to 22.3). Seven patients accepted neoadjuvant chemotherapy and there was no significant morbidity among them,with a resection rate of R0 of 5/7. Conclusion:DP-CAR is safe and feasible for selective cases,which could be more valuable in improving long-term survival when combined with (neo) adjuvant therapy.

Objective:To investigate the clinical efficacy of distal pancreatectomy with celiac axis resection(DP-CAR).Methods:A total of 89 consecutive patients (50 males and 39 females) who were diagnosed with pancreatic body cancer and underwent DP-CAR in Pancreas Center,First Affiliated Hospital of Nanjing Medical University between September 2013 and June 2022 were retrospectively reviewed. There were 50 males and 39 females,with age( M(IQR)) of 63(12) years(range:43 to 81 years). Perioperative parameters,pathology results and follow-up data of these patients were analyzed, χ2 or Fisher′s test for categorical data while the Wilcoxon test for quantitative data. Survival results were estimated by the Kaplan-Meier survival method. Results:Among 89 cases,cases combined with portal vein-superior mesenteric vein or organ resection accounted for 22.5% (20/89) and 42.7% (38/89),respectively. The operative time,blood loss and postoperative hospital stay were 270 (110) minutes,300 (300) ml and 13 (10) days,respectively. The overall morbidity rate was 67.4% (60/89) while the major morbidity was 11.2% (10/89). The increase rate in transient liver enzymes was 42.7% (38/89),3.4% (3/89) for liver failure,53.9% (48/89) for clinically relevant postoperative pancreatic fistula,1.1% (1/89) for bile leak,3.4% (3/89) for chylous leak of grade B and C,11.2% (10/89) for abdominal infection,9.0% (8/89) for postoperative hemorrhage of grade B and C,4.5% (4/89) for delayed gastric emptying,6.7% (6/89) for deep vein thrombosis,3.4% (3/89) for reoperation,4.5% (4/89)for hospital mortality,7.9% (7/89) for 90-day mortality. The pathological type was pancreatic cancer for all 89 cases and pancreatic ductal adenocarcinoma made up 92.1% (82/89). The tumor size was 4.8(2.0) cm, ranging from 1.5 to 12.0 cm. The number of lymph nodes harvested was 14 (13)(range:2 to 33),with a positive lymph node rate of 13.0% (24.0%). The resection R0 rate was 30.0% (24/80) and the R1 (<1 mm) rate was 58.8% (47/80). The median overall survival time was 21.3 months (95% CI: 15.6 to 24.3) and the median disease-free survival time was 19.1 months (95% CI: 11.7 to 25.1). The overall survival at 1-year and 2-year were 69.60% and 39.52%. The median survival time of 58 patients with adjuvant chemotherapy was 24.3 months (95% CI: 17.8 to 32.3) while that of 13 patients without any kind of adjuvant therapy was 8.4 months (95% CI: 7.3 to 22.3). Seven patients accepted neoadjuvant chemotherapy and there was no significant morbidity among them,with a resection rate of R0 of 5/7. Conclusion:DP-CAR is safe and feasible for selective cases,which could be more valuable in improving long-term survival when combined with (neo) adjuvant therapy.

Objective:To evaluate the value of left-sided uncinate process first approach in pancreaticoduodenectomy.Methods:The clinical data of 152 patients who underwent the left-sided uncinate process first approach during pancreaticoduodenectomy at Pancreas Center, the First Affiliated Hospital of Nanjing Medical University from January 2020 to December 2020 were analyzed retrospectively. There were 64 females and 88 males,with age( M( Q R)) of 62.0(14.7)years(range:16.0 to 84.0 years). The clinical date of 117 patients who underwent pancreaticoduodenectomy without using left-sided uncinate process first approach in the same period was selected as the control group,including 65 females and 52 males,with age of 64.0(13.0) years(range:13.0 to 84.0 years). Fisher exact probability method and t test were used to compare the data between the two groups,rank sum test was used for comparison of continuous variables between the two groups. Results:Pancreaticoduodenectomy was successfully performed in 152 patients in left-sided uncinate process first approach group. The operation time was 222.5(77.0) minutes(range:117.0 to 480.0 minutes),the time of uncinate process resection from left-side(the time from jejunum dissection to complete dissociation of the uncinate process) was 11.0(4.5) minutes(range:7.5 to 20.0 minutes),the time of pancreatic head resection (the time from jejunum dissection to pancreaticoduodenal specimen removal) was 26.0(8.5) minutes(range:20.0 to 41.0 minutes),the intraoperative blood loss was 200(150) ml(range:50 to 800 ml),and the intraoperative blood transfusion rate was 9.2% (14/152). Postoperative conditions:The postoperative hospital stay was 12 (9) d(range:6 to 55 d),the overall incidence of postoperative complications was 59.9%(91/152),and there was no perioperative death. Pathological results:The R0 resection rate of periampullary malignant tumor was 64.3%(77/112),with negative rate of uncinate process margin was 91.1%(102/112). The R0 resection rate of pancreatic ductal adenocarcinoma was 46.9%,with negative rate of uncinate process margin was 89.1%(57/64). Compared with the non-left-sided uncinate process first approach group(222.5(77.0) minutes, 9.2%(14/152)),the left-sided uncinate process first approach group had shorter operation time(246.0(94.0) minutes) ( Z=3.964, P<0.01),less intraoperative blood loss (18.8%(22/117))( Z=4.843, P<0.01),and lower intraoperative blood transfusion rate(χ2=5.248, P=0.029). However,there were no significant differences between two groups in postoperative hospital stay( Z=1.682, P=0.093),postoperative overall complications( P=0.549),R0 resection rate of periampullary malignant tumor(χ2=2.012, P=0.156),and negative rate of uncinate process margin(χ2=2.108, P=0.147). Conclusions:The "left-sided uncinate process first approach" could completely resect uncinate process under a direct vision,especially when the uncinate process was behind the superior mesenteric artery or beyond the left lateral margin of the superior mesenteric artery. The "left-sided uncinate process first approach" might increase the negative rate of uncinate process margin and R0 resection rate for periampullary malignant tumor.

Although the incidence of pancreatic cancer is increasing, due to the develop-ment of surgery, oncology and other disciplines, its prognosis has improved. The anatomical neigh-boring of pancreatic neck cancer is complex and there is no consensus in surgical decision-making such as the extension of pancreatic resection, handling of arteriovenous invasion, and lymph node dissection, which hampered the development of safety and standardization for radical surgery. The authors review the key points of surgical decision-making in radical surgery for pancreatic neck cancer based on evidence-based medicine and clinical experience.

Objective:To investigate the clinical efficacy of pancreaticoduodenectomy (PD) for periampullary diseases.Methods:The retrospective cohort study was conducted. The clinico-pathological data of 2 019 patients with periampullary diseases who underwent PD in the First Affiliated Hospital of Nanjing Medical University from January 2016 to December 2020 were collected. There were 1 193 males and 826 females, aged 63(15) years. Observation indicators: (1) surgical situations; (2) postoperative conditions; (3) postoperative pathological examinations; (4) prognosis of patients with periampullary carcinoma. Regular follow-up was conducted by telephone interview and outpatient examination once every 3 months within the postoperative first year and once every 6 months thereafter to detect the survival of patients with periampullary carcinoma. The follow-up was up to December 2021. Measurement data with skewed distribution were represented as M(IQR) or M(range), and comparison between groups was analyzed using the rank sum test. Count data were described as absolute numbers and (or) percentages, and comparison between groups was analyzed by the chi-square test or Fisher exact probability. Kaplan-Meier method was used to draw survival curves and calculate survival rates, and Log-Rank test was used to conduct survival analysis. Results:(1) Surgical situations: of 2 019 patients, 1 116 cases were admitted from 2016-2018 and 903 cases were admitted from 2019-2020. There were 1 866 cases undergoing open PD and 153 cases undergoing laparoscopic or robot-assisted PD. There were 1 049 cases under-going standard PD and 970 cases undergoing pylorus-preserved PD. There were 215 cases combined with portal mesenteric vein resection, 3 cases combined with arterial resection. The operation time of 2 019 patients was 255(104)minutes and the volume of intraoperative blood loss was 250(200)mL. The intraoperative blood transfusion rate was 31.401%(623/1 984), with the blood transfusion data of 35 cases missing. The proportions of pylorus-preservation, combination with portal mesenteric vein resection, intraoperative blood transfusion were 585 cases(52.419%), 97cases(8.692%), 384 cases(34.941%) for patients admitted in 2016-2018, versus 385 cases(42.636%), 118 cases(13.068%), 239 cases(27.006%) for patients admitted in 2019-2020, showing significant differences between them ( χ2=19.14,10.05,14.33, P<0.05). (2) Postoperative conditions: the duration of postoperative hospital stay of 2 019 patients was 13 (10) days. One of 2 019 patients lacked the data of postopera-tive complications. The overall postoperative complication rate was 45.292%(914/2 018), of which the incidence rate of grade B or C pancreatic fistula was 23.439%(473/2 018), the rate of grade B or C hemorrhage was 8.127%(164/2 018), the rate of grade B or C delayed gastric emptying was 15.312%(309/2 018), the rate of biliary fistula was 2.428%(49/2 018) and the rate of abdominal infection was 12.884%(260/2 018). The reoperation rate of 2 019 patients was 1.932%(39/2 019), the in-hospital mortality was 0.644%(13/2 019), the postoperative 30-day mortality was 1.238%(25/2 019), and the postoperative 90-day mortality was 2.675%(54/2 019). There were 541 cases(48.477%) with overall postoperative complications, 109 cases(9.767%) with grade B or C hemorr-hage, 208 cases(18.638%) with grade B or C delayed gastric emptying , 172 cases(15.412%) with abdominal infection, 39 cases(3.495%) with postoperative 90-day mortality of 1 116 patients admitted in 2016-2018. The above indicators were 373 cases(41.353%), 55 cases(6.098%), 101 cases(11.197%), 88 cases(9.756%), 15 cases(1.661%) of 902 patients admitted in 2019-2020, respectively. There were significant differences in the above indicators between them( χ2=10.22, 9.00, 21.30, 14.22, 6.45 , P<0.05). The in-hospital mortality occurred to 11 patients(0.986%) of 1 116 patients admitted in 2016-2018 and to 2 cases(0.221%) of 903 patients admitted in 2019-2020, showing a significant difference between them ( P<0.05). (3) Postoperative pathological examinations. Disease area of 2 019 patients reported in postoperative pathological examinations: there were 1 346 cases(66.667%) with lesions in pancreas, including 1 023 cases of carcinoma (76.003%) and 323 cases(23.997%) of benign diseases or low potential malignancy. There were 250 cases(12.382%) with lesions in duodenal papilla, including 225 cases of carcinoma (90.000%) and 25 cases(10.000%) of benign diseases or low potential malignancy. There were 174 cases(8.618%) with lesions in bile duct, including 156 cases of carcinoma (89.655%) and 18 cases(10.345%) of benign diseases or low potential malignancy. There were 140 cases(6.934%) with lesions in ampulla, including 134 cases of carcinoma (95.714%) and 6 cases(4.286%) of benign diseases or low potential malignancy. There were 91 cases(4.507%) with lesions in duodenum, including 52 cases of carcinoma (57.143%) and 39 cases(42.857%) of benign diseases or low potential malignancy. There were 18 cases(0.892%) with carcinoma in other sites. Postoperative pathological examination showed carcinoma in 1 608 cases(79.643%), benign diseases or low potential malignancy in 411 cases(20.357%). The histological types of 1 608 patients with carcinoma included adenocarcinoma in 1 447 cases (89.988%), intra-ductal papillary mucinous carcinoma in 37 cases(2.301%), adenosquamous carcinoma in 35 cases(2.177%), adenocarcinoma with other cancerous components in 29 cases(1.803%), neuroendocrine carcinoma in 18 cases(1.119%), squamous carcinoma in 1 case (0.062%), and other histological malignancies in 41 cases(2.550%). The histological types of 411 patients with benign or low poten-tial malignancy included intraductal papillary mucinous neoplasm in 107 cases (26.034%), chronic or autoimmune inflammatory disease in 62 cases(15.085%), neuroendocrine tumor in 58 cases(14.112%), pancreatic serous cystadenoma in 52 cases(12.652%), pancreatic solid pseudopapillary tumor in 36 cases(8.759%), gastrointestinal stromal tumor in 29 cases(7.056%), villous ductal adenoma in 20 cases(4.866%), pancreatic mucinous cystadenoma in 2 cases(0.487%), pancreatic or duodenal trauma in 2 cases(0.487%) and other histological types in 43 cases(10.462%). (4) Prognosis of patients with periampullary carcinoma. Results of survival analysis of 1 590 patients with main locations of periampullary carcinoma showed that of 1 023 patients with pancreatic cancer, 969 cases were followed up for 3.0-69.6 months, with a median follow-up time of 30.9 months. The median overall survival time, 1-year, 3-year and 5-year survival rates of pancreatic cancer patients were 19.5 months [95% confidence interval ( CI) as 18.0-21.2 months], 74.28%, 29.22% and 17.92%. Of 225 patients with duodenal papillary cancer, 185 cases were followed up for 3.0-68.9 months, with a median follow-up time of 36.7 months. The median overall survival time, 1-year, 3-year and 5-year survival rates were unreached, 94.92%, 78.87% and 66.94%. Of 156 patients with distal bile duct cancer, 110 cases were followed up for 3.0-69.5 months, with a median follow-up time of 25.9 months. The median overall survival time, 1-year, 3-year and 5-year survival rates were 50.6 months (95% CI as 31.4 to not reached), 90.37%, 56.11% and 48.84%. Of 134 patients with ampullary cancer, 100 cases were followed up for 3.0-67.8 months, with a median follow-up time of 28.1 months. The median overall survival time, 1-year, 3-year and 5-year survival rates were 62.4 months (95% CI as 37.8 months to not reached), 90.57%, 64.98% and 62.22%. Of 52 patients with duodenal cancer, 38 cases were followed up for 3.0-69.5 months, with a median follow-up time of 26.2 months. The median overall survival time, 1-year, 3-year and 5-year survival rates were 52.0 months (95% CI as 30.6 months to not reached), 93.75%, 62.24% and 40.01%.There was a significant difference in overall survival after PD between patients with different locations of periampullary malignancies ( χ2=163.76, P<0.05). Conclusions:PD is safe and feasible in a high-volume pancreas center, but the incidence of overall postoperative complications remains high. With the increase of PD volume, the incidence of overall postoperative complications has significantly decreased. There is a significant difference in overall survival time after PD among patients with different locations of periampullary malignancies. The 5-year survival rate after PD for duodenal papillary cancer, ampullary cancer, duodenal cancer and distal bile duct cancer is relatively high, whereas for pancreatic cancer is low.