【Abstract】 Objective To investigate the origin, prevention and treatment of postoperative complications and death rate after pancreaticoduodenectomy （PD）. Methods Retrospective study on the clinical materials of complications and death rate was done on 106 cases of PD performed in our hospital during July 1985 to December 2002. Results In this group, 37 cases (34.91%) had postoperative complications, and the incidence rate of severe complications was 19.81% (21/106), the death rate was 10.38% (11/106). Compared between the two groups with preoperative bilirubin gt;342 μmol/L and ≤342 μmol/L, the incidence of total complications increased evidently (P＜0.05), and the bleeding amount，infusion amount and operation time in those with complications or dead ones were evidently higher than those without complications (P＜0.05). Conclusion The safty and resectability of PD has improved evidently in recent years but good skills, careful operation, the experience of the operatior and careful perioperative treatment and nursing are of crucial importance to reduce the complications and death rate.
【Abstract】 Objective To investigate the preoperative and intraoperative evaluation and the operative methods of pancreaticoduodenectomy combined with resection of involved blood vessel. Methods The materials of 226 cases of carcinoma of head of pancreas, which were collected from January 2002 to June 2005, were assessed according to the T-grade evaluative method and were summarized into a more accurate method of preoperative and intraoperative evaluation and an operative method of the pancreaticoduodenectomy combined with vascular resection. The statistical analyses were performed by SAS 6.12 Stat software package with χ2 test for categorical data. Results The coincidence of every preoperative evaluations on pancreaticoduodenectomy with T-grade method were more than 85.00% and the accurate rate of evaluation on relationship between tumor and blood vessel was 80.97%. As the degree of T-grade increased, the proportion of requiring pancreaticoduodenectomy combined with vascular resection increased significantly as well (P＜0.001). All operations were smoothly done and the reconstruction of removed vascular took 12 to 46 minutes. The complication rate and the mortality rate of perioperative period were 19.61% and 1.96%, respectively. Conclusion With more practice, experience and the advanced imaging technology, it is possible for professional surgeon to evaluate more accurately on the pancreaticoduodenectomy combined with vascular resection by using the preoperative T-grade method. T-grade evaluative method is useful for operator to assess the indication accurately and to design the operative schedule reasonably. Accurate operational evaluation and careful operation are also important for pancreatoduodenectomy combined vascular resection.
ObjectiveTo introduce the method of laparoscopic pancreaticoduodenectomy through an arterial approach, and to evaluate the clinical value of this technique. MethodsThe clinical data of 19 patients with periampullary carcinoma, distal bile duct cancer, and early-stage pancreatic head carcinoma that underwent laparoscopic pancreaticoduodenectomy through an arterial approach in the Department of Hepatobiliary and Pancreatic Surgery, Sun Yat-sen Memorial Hospital between September 2010 and July 2013 were retrospectively analyzed. The patients were followed-up until February 28, 2014. ResultsLaparoscopic pancreaticoduodenectomy were successfully performed in all 19 cases, there were no need to convert to open surgery. Open reconstruction was performed in 2 cases, and 17 cases underwent total laparoscopic reconstruction of the digestive tract. The duration of the operations ranged from 5-10.5 h(mean 6.3 h), and the intra-operative blood loss ranged from 170-430 mL(mean 250 mL). Post-surgical pathology detected a mean number of 13.7 lymph(9-21) nodes in all patients. No deaths occurred during the perioperative period. Complications were observed in 42.1%(8/19) of the subjects, including 5 cases with pancreatic fistula, 1 case with bile leak, 1 case with gastric emptying disorder, and 1 case with a gastroduodenal artery aneurysm. The mean length of hospital stay was 10.7 d(7-19 d). The mean followed-up period was 7.5 months(2-28 months), there were 6 patients died of tumor metastasis or recurrence during the followed-up. ConclusionLaparoscopic pancreaticoduodenectomy through an arterial approach simplifies pancreaticoduodenectomy and lymph node dissection procedures, and can completely remove lymph nodes.
ObjectiveTo evaluate the postoperative complications after pancreaticoduodenectomy with modified triple-layer(MTL) duct-to-mucosa pancreaticojejunostomy and with resection of jejunal serosa, analyse the risk factors of pancreatic fistula, and compare effects with two-layer(TL) duct-to-mucosa pancreaticojejunostomy. MethodsData on 184 consecutive patients who underwent the two methods of pancreaticojejunostomy during standard PD between January 1, 2010 and January 31, 2013 were collected retrospectively. The risk factors of pancreatic fistula were investigated by using univariate and multivariate analyses. ResultsA total of 88 patients received TL and 96 underwent MTL. Rate of pancreatic fistula for the entire cohort was 8.2%(15/184). There were 11 fistulas(12.5%) in the TL group and four fistulas(4.2%) in the MTL group(P=0.039). Body mass index, pancreatic texture, pancreatic duct diameter, and methods of pancreaticojejunostomy had significant effects on the formation of pancreatic fistula on univariate analysis. Multivariate analysis showed that pancreatic duct diameter less than 3 mm and TL were the significant risk factors of pancreatic fistula. ConclusionsMTL technique effectively reduced the pancreatic fistula rate after PD in comparison with TL, especially in patients with pancreatic duct diameter less than 3 mm.
ObjectiveTo assess the safety for removing nasogastric tube(NGT)within postoperative 24 h in Whipple pancreaticoduodenectomy (PD)patients. MethodsThe clinical data of 310 patients performed classic Whipple PD from January 2008 to March 2013 in this hospital were analyzed retrospectively. The patients were divided into early (≤24 h after operation)removing NGT group and late( > 24 h after operation)removing NGT group according to the time of NGT duration. The ratio of NGT reinsertion, time of solid diet tolerance, hospital stay, mortality, and major complications associated with PD were compared between two groups. Results①The demography and preoperative comorbidities characteristics were similar(P > 0.05).②There was no statistical difference of ratio of NGT reinsertion between two groups(P=0.450).③The differences of rates of major complications associated with PD and mortality were not statistically different(P > 0.05)by univariate analysis, but the rate of total complications in the early removing NGT group was significantly lower than that in the late removing NGT group (P=0.014)by multivariate analysis.④The average time of solid diet tolerance(P=0.013)and average hospital stay(P < 0.001)in the early removing NGT group were significantly shorter than those in the late removing NGT group. ConclusionFor patients comfort, NGT following PD should be removed as early as possible even immediately after extubation for selective patients.
Objective To investigate the application value of the binding pancreaticogastrostomy in pancreatico-duodenectomy. Methods The clinical data of 13 patients that performed pancreaticoduodenectomy with binding pancr-eaticogastrostomy from Jan. 2010 to Mar. 2013 in our hospital were retrospectively analyzed. The incidence of postoper-ative complications were counted. Results There was 1 patient with pancreatic stump bleeding after operation， and then recovered after conservative treatment. There was no patient with pancreatic fistula， bile fistula， delayed gastric empt-ying， and other complications after operation in whole group. Peritoneal fluid and amylase level in peritoneal fluid were gradually reduced or degraded after operation. The gastrointestinal function was recovered better. All patients were compl-etely cured. Conclusion The binding pancreaticogastrostomy in pancreaticoduodenectomy has its own unique advantage.It could be reduce the incidence of pancreatic fistula in postoperative patients by using binding pancreaticogastrostomy reasonably.
Objective To analyze the difference in the incidence of postoperative pancreatic leakage and anasto-motic bleeding complications in various methods of pancreaticojejunostomy after pancreaticoduodenectomy （PD）. Methods The clinical data of 526 patients underwent pancreaticojejunostomy from January 2008 to September 2012 in this hospital were analyzed retrospectively. End-to-side “pancreatic duct to jejunum mucosa-to-mucosa” anastomosis （abbreviation：mucosa-to-mucosa anastomosis） was performed in 359 patients， which contained 149 patients with internal drainage， 130 patients with external drainage， and 80 patients with no drainage. End-to-side invaginated anastomosis was performedin 165 patients without drainage. In addition， side-to-side anastomosis was performed in 2 patients without drainage.Results There were 34 cases （6.46%） of pancreatic leakage， 8 cases （1.52%） of anastomotic bleeding in pancreaticoje-junostomy， and 32 cases of death （6.08%）. ① The pancreatic leakage rate of mucosa-to-mucosa anastomosis was signi-ficantly lower than that of end-to-side invaginated anastomosis 〔4.18% （15/359） versus 11.52% （19/165）， χ2=10.029， P=0.002〕. There was no significant difference of the anastomotic bleeding incidence between mucosa-to-mucosa anasto-mosis and end-to-side invaginated anastomosis 〔1.67% （6/359） versus 1.21% （2/165）， χ2=0.159， P=0.691〕. ② In the mucosa-to-mucosa anastomosis group， the pancreatic leakage rates in the ones with internal drainage and external drainage were lower than those in the ones without drainage， respectively （2.68% （4/149） versus 11.25% （9/80）， χ2=7.132， P=0.008；1.54% （2/130） versus 11.25% （9/80）， χ2=9.410， P=0.002）；which was no significant difference between the ones with internal drainage and external drainage 〔2.68% （4/149） versus 1.54% （2/130）， χ2=0.433， P=0.510〕. But there were no significant differences for both the pancreatic leakage 〔2.68% （4/149） versus 1.54% （2/130）， χ2=0.433， P=0.510〕and anastomotic bleeding incidence 〔2.68% （4/149） versus 1.54% （2/130）， χ2=0.433， P=0.510〕 between the ones with internal drainage and external drainage. Conclusions Mucosa-to-mucosa anastomosis has a lower pancreatic leakage incidence as compared with end-to-side invaginated anastomosis. However， there is no significant difference of the anast-omotic bleeding incidence. Internal or external drainage could reduce the incidence of pancreatic leakage， but have no obvious effect to the anastomotic bleeding incidence.
Objective To explore the situation and prevention of pancreaticoduodenectomy perioperative complications. Methods The clinical data of 111 cases of pancreaticoduodenectomy were retrospectively analyzed， and the possible factor of complications was analyzed. Results There were postoperative complications in 48 patients （43.2%）， which one kind complication occurred in 25 cases， two kinds in 15 cases， and three kinds or more in 8 cases. Four cases （3.6%） died after operation. Conclusions Pancreaticoduodenectomy is a higher risk surgery in abdominal operation. Strengthen perioperative prevention and treatment are important measures to reduce morbidity and mortality after pancreaticoduodenectomy.
Objective To evaluate the operative indication and results of pancreaticogastrostomy following pancreaticoduodenectomy. Methods A retrospective study was carried out on the cases of pancreaticoduodenectomy following pancreaticogastrostomy from Aug. 2005 to Feb. 2008 in Shanghai Tongji Hospital. Results During this period, 38 cases had undergone pancreaticogastrostomy with pancreaticoduodenectomy. The median operative time was (352.1±78.3) min. The median intraoperative blood transfusion was (911.3±601.4) ml. The median postoperative length of stay was (26.2±12.1) d. Postoperative morbidity was 21.1% (8/38) with no operative death. Pancreatic anastomotic leakage occurred in 1 patient. Delayed gastric emptying occurred in 2 patients. Incision infection occurred in 2 patients. Abdominal fluid collection occurred in 1 patient and pulmonary infection occurred in 2 patients. All of the complications were treated conservatively. Conclusion Pancreaticogastrostomy is a safer drainage procedure for the pancreatic stump after pancreaticoduodenectomy.
ObjectiveTo evaluate the predictive value of pancreatic duct diameter and pancreatic gland thickness measured using preoperative CT imaging on pancreatic fistula(PF)following pancreaticoduodenectomy (PD). MethodsOne hundred and fifty-one patients who underwent PD consecutively from January 2013 to April 2014 were reviewed retrospectively. Associations between the gender, age and the pancreatic duct diameter and pancreatic gland thickness from preoperative CT imaging and PF were analyzed. The diagnostic values of the pancreatic duct diameter and pancreatic gland thickness in patients with PF were evaluate by receiver operating characteristic (ROC) analysis. The reliability analysis was done for the pancreatic duct diameter and pancreatic gland thickness by using the intraclass correlation coefficient (ICC). The Spearman rank correlation analysis was done between the pancreatic duct diameter and pancreatic gland thickness. Results①PF occurred in 46 cases (30.1%).②The gender and age were not associated with PF (Gender: χ2=1.698, P=0.193; Age: χ2=0.016, P=0.900). The pancreatic duct diameter and pancreatic gland thickness were associated with PF (Pancreatic duct diameter: OR=0.275, 95% CI 0.164-0.461, P=0.000; Pancreatic gland thickness: OR=1.319, 95% CI 1.163-1.496, P=0.000).③There was no correlation between the pancreatic duct diameter and the pancreatic gland thickness (rs=-0.120, P=0.145).④The area under curve of ROC was 0.814 (95% CI 0.745-0.883, P < 0.001) for the pancreatic duct diameter in predicting the PF, the sensitivity and specificity was 68.6% and 78.3% respectively when the best critical value was 3.5 mm. The area under curve of ROC was 0.762 (95% CI 0.674-0.849, P < 0.001) for the pancreatic gland thickness in predicting PF, the sensitivity and specificity was 63.0% and 85.7% respectively when the best critical value was 31 mm.⑤The ICC of the pancreatic duct diameter and pancreatic gland thickness was 0.984 and 0.992 respectively by two medical diagnostic measurement. ConclusionPancreatic duct diameter and pancreatic gland thickness measured using preoperative CT imaging might be useful in predicting PF following PD.