In recent years, the mortality rate of pancreaticoduodenectomy has decreased to less than 5% worldwide, but the incidence of its important complication, pancreatic fistula, has not decreased significantly and is still maintained between 5% and 40% depending on the definition of postoperative pancreatic fistula, and how to manage the pancreatic stump after pancreaticoduodenectomy remains an urgent clinical problem. Pancreatic-intestinal anastomosis is the classical method of postoperative reconstruction, and many techniques have been explored, such as pancreatic duct ligation, bioprotein adhesive sealing of the pancreatic-intestinal anastomosis, pancreaticogastric anastomosis, internal drainage of the pancreatic duct, sleeve-in anastomosis, bundled anastomosis, separation of pancreatic-intestinal and biliary-intestinal anastomosis, and prophylactic application of growth inhibitors, but randomized prospective studies have not found significant reduction in the incidence of pancreatic fistula, and only external drainage of the pancreatic duct support is considered to be Only the external drainage of pancreatic duct support is considered to be an important part of reducing the incidence of pancreatic fistula. In recent years, pancreatic duct jejunal mucosal anastomosis has received much attention, but the findings are inconsistent, and only a very few publications are randomized prospective comparisons of the effect of surgical technique on the incidence of pancreatic fistula. In view of this, this randomized prospective study was designed to verify whether pancreatic ductojejunal mucosal anastomosis can significantly reduce the incidence of pancreatic fistula after pancreaticoduodenectomy.
1. Data and methods
(1) Clinical data
From October 2006 to December 2008, a total of 69 pancreaticoduodenectomies were performed electively in our hospital. 64 of these patients had soft pancreatic stumps, which were pathologically diagnosed as non-fibrotic and soft, and were at high risk of postoperative pancreatic fistula, and were randomly grouped into pancreatic duct-jejunostomy group (Group A) and pancreatic-enteric sleeve anastomosis group (Group B), with 32 patients in each group. patients. Another 4 patients (5.8%) had chronic pancreatitis confirmed by postoperative pathology, and one case had biliary peritonitis due to accidental removal of the gallbladder puncture tube 2 days before surgery, which was excluded from the experiment despite definitive surgery.
(2) Surgical methods
All patients underwent conventional pancreaticoduodenectomy, including two cases of partial resection and repair of the lateral wall of the portal vein, one case of portal vein resection with end-to-end anastomosis, and one case of total visceral inversion with reverse pancreaticoduodenectomy, and the sequence of digestive tract reconstruction was performed by the Child method. The pancreatic-intestinal anastomosis was decided by a random number table, and a uniform standardized operation was used. The pancreatic duct jejunal mucosal end-lateral anastomosis was closed with 4-0 prolene suture in interrupted double layers; the pancreatic-intestinal sleeve anastomosis had end-lateral anastomosis in 20 cases and end-end sleeve anastomosis in 12 cases, depending entirely on the size of the pancreatic stump. External pancreatic duct drainage was used in 58 patients (90.6%), and there was no significant difference between the two groups (28 patients in group A; 30 patients in group B), and internal pancreatic duct drainage was used in another 5 patients (7.8%), and all pancreatic duct stumps were inserted with the largest diameter catheter possible to facilitate adequate drainage, and only one patient was not supported for drainage because the main pancreatic duct was not found intraoperatively. After bile-intestinal anastomosis, external T-tube drainage was placed in the bile duct, and jejunal nutrition tubes were routinely placed for postoperative enteral nutrition, and drainage tubes were placed near the pancreatic-intestinal and bile-intestinal anastomoses for drainage.
(3) Definition of pancreatic fistula
Since the diagnostic criteria and classification of pancreatic fistula after pancreaticoduodenectomy are still controversial, this paper adopts the International Study Group on Pancreatic Fistula (ISGPF) 2005 criteria, pancreatic fistula is defined as postoperative peripancreatic drainage of more than 30 ml/d and an amylase level more than three times the normal serum value. Pancreatic fistulas were further subdivided into two types: light (type I), transient or asymptomatic pancreatic fistulas with only elevated amylase in the drainage fluid, which can heal spontaneously with non-surgical treatment; and heavy (type II), pancreatic fistulas with significant clinical symptoms, complicated by sepsis or abdominal abscesses, requiring puncture drainage or even surgical intervention.
(4), Statistical methods
SPSS13.0 software was applied for data processing. The measured continuous data were expressed as mean ± standard deviation (±s), and the t-test was used to compare the means of two groups, and the number of cases and corresponding proportions of each category were given for the count data, and the Fisher exact probability test was used for comparison between groups. p≤0.05 was considered a statistically significant difference.
2.Results
(1), Basic situation
All 64 patients underwent successful pancreaticoduodenectomy, and there were no significant differences in age, gender, clinical symptoms, preoperative comorbidities, preoperative biliary drainage, pathological diagnosis and pancreatic duct diameter between the two groups, as shown in Table 1. preoperative biliary drainage was basically not possible in this group, and the main surgical pointers were pancreatic head cancer and peri-potbelly cancer.
(2), complications and prognosis
There were 21 (32.8%) surgical complications in 64 patients, among which a total of 8 (12.5%) patients developed pancreatic fistula, only 2 (6.25%) in the pancreatic duct mucosal anastomosis group and 6 (18.8%) in the pancreatic-enteric sleeve anastomosis group, there was no significant difference between the two (P<0.05), but 4 patients in the pancreatic-enteric sleeve anastomosis group developed clinically significant type II pancreatic fistula, compared with the pancreatic duct mucosal anastomosis group compared to the pancreatic duct group (P=0.040, Table 2.). Other complications included biliary fistula, abdominal abscess, intra-abdominal hemorrhage, impaired gastric emptying, gastrointestinal hemorrhage, pulmonary infection, and incisional infection in a total of 17 cases (25%), with no significant difference between the two groups. Two patients (3.1%) in this group were reoperated, both in the pancreatic-intestinal sleeve anastomosis group. One case was cured after non-surgical treatment for postoperative abdominal hemorrhage, which turned into abdominal abscess, and the other case was also cured after reoperative treatment for postoperative abdominal hemorrhage and emergency reoperative exploration with gauze tamponade, but the symptoms were still uncontrollable after surgery, and the family gave up treatment and was discharged automatically. The other death was seen in the pancreatic duct mucosal anastomosis group, where the patient had undergone cholecystectomy in an outside hospital 5 months ago and developed jaundice after surgery, which rapidly increased progressively, and was found to have an occupied pancreatic head on admission.
(3) Transition of pancreatic fistula
All 8 patients with pancreatic fistula were diagnosed by postoperative pancreatic-enteric anastomosis or biliary-enteric anastomosis drainage and quantitative amylase measurement of abdominal drainage fluid. 4 cases of simple type I pancreatic fistula were spontaneously cured after non-operative treatment, and 4 cases of type II pancreatic fistula with combined abdominal infection and/or sepsis were also cured after intensive drainage (abscess puncture drainage and active flushing drainage with negative pressure double cannula), nutritional support, application of growth inhibitors, anti-infection and other symptomatic supportive treatments. It was cured.
3. Discussion
Pancreatic fistula is a common complication and an important cause of death after pancreaticoduodenectomy, and in order to reduce perioperative mortality and surgical complications, many strategies and methods have been adopted to improve the traditional pancreatic-intestinal anastomosis, including the use of some prophylactic drugs, regional concentration of surgical cases, and the choice of surgical techniques, etc., but no consensus has been reached, among which the surgical techniques are the most explored and the most controversial. Prospective randomized studies are few and far between. Pancreatic-enteric anastomosis is still the most widely used mode of GI reconstruction, and it is more necessary to seek technical progress in order to minimize the incidence of pancreatic fistula, and we designed this prospective randomized study to verify whether pancreatic-jejunal mucosal anastomosis can effectively reduce the incidence of pancreatic fistula after pancreaticoduodenectomy.
The pancreatic duct mucosal jejunostomy was first pioneered by Varco in 1945 and has been highly recommended recently. It is believed that the anterior and posterior edges of the pancreatic section are sutured together, and the tight fit between the pancreatic section and the intestinal wall makes the anastomotic surface free from fluid accumulation and fast fit, thus effectively preventing bleeding and leakage of pancreatic fluid from the pancreatic section; the direct anastomosis between the intestinal mucosa and the pancreatic duct not only promotes rapid healing, but also maximizes the patency of the pancreatic-enteric However, most of these findings were found in retrospective analyses, and an earlier prospective randomized study in piglets showed that pancreatic ductal mucosal anastomosis was significantly superior to pancreatic ductal sleeve anastomosis, but the only prospective randomized clinical study found that pancreatic ductal mucosal anastomosis did not significantly reduce the incidence of pancreatic fistula and other surgical complications compared with conventional pancreatic-enteric sleeve anastomosis, but this study did not fully standardize the technique of pancreatic-enteric anastomosis. Only 57% of patients had external pancreatic duct drainage, which is one of the important means to reduce the incidence of pancreatic fistula, and some patients who did not have pancreatic duct drainage may have increased the incidence of pancreatic fistula as a result, thus the advantages of pancreatic mucosal anastomosis may not be fully exploited, which may lead to bias in the experimental results.
In this study, there were no significant differences between the two groups in terms of underlying disease, type of pathology, preoperative biliary drainage, pancreatic duct diameter, and pancreatic duct drainage, and the surgical approach was standardized and controlled, and the texture of the residual pancreas in all the cases enrolled was soft pancreas, because literature studies have shown that the texture of the residual pancreas is significantly associated with the incidence of pancreatic fistula, while the incidence of pancreatic-enteric anastomotic fistula is greatly reduced in patients with chronic pancreatitis. Also, we did not apply growth inhibitors prophylactically to further reduce experimental bias. In view of this, the two groups of cases in this study were well homogenous and comparable. We found no significant difference in the rate of surgical complications and overall pancreatic fistula incidence between the two groups of cases, and for type II pancreatic fistula with severe clinical symptoms, there was no significant difference between the two groups for comparison (0 vs. 12.5%, P=0.057).
The overall incidence of pancreatic fistula was 12.5% in our group, 6.25% in the group with pancreatic ductal mucosal anastomosis and 18.75% in the group with pancreatic-enteric sleeve anastomosis, which is similar to some of the results in the literature. However, the incidence of pancreatic fistula in the literature is only 2% to 8%, and even pancreatic fistula was completely avoided. However, the definition of pancreatic fistula in the above-mentioned studies is generally too strict, requiring higher amylase concentration in the abdominal drainage fluid, more volume of abdominal drainage fluid, or longer time to elevation of amylase in the abdominal drainage fluid. The definition of pancreatic fistula after pancreaticoduodenectomy is currently debated, and a recent study showed that the incidence of pancreatic fistula varied in the same group of cases due to different definitions of pancreatic fistula. 242 patients with pancreatic-enteric anastomosis, if pancreatic fistula was defined as “amylase-rich drainage of more than 10 ml from the fifth postoperative day”, the incidence of pancreatic fistula was 28.5 The incidence of pancreatic fistula was only 9.9% if it was defined as “more than 50 ml of amylase-rich drainage fluid from the eleventh postoperative day”. In this paper, we adopt the diagnostic criteria of the International Study Group for Pancreatic Fistula (ISGPF) in 2005 and do not emphasize the need for imaging. However, the definition and staging of pancreatic fistula after pancreaticoduodenectomy in China are still few and need to be further investigated in order to explore the best management of the pancreatic stump.
In conclusion, this single-center prospective randomized study showed that pancreatic ductal mucosal anastomosis failed to significantly reduce the incidence of pancreatic fistula after pancreaticoduodenectomy compared with pancreatic-enteric sleeve anastomosis, but the sample size was small, and further in-depth studies on the effects of pancreatic ductal mucosal anastomosis on surgical complications, pancreatic fistula incidence, reoperation rate, and postoperative mortality after pancreaticoduodenectomy are necessary.