The Effects Of Negative Pressure Drainage With Different Modes On Postoperative Complications After Pancreaticoduodenectomy

Objective:Pancreaticoduodenectomy is an effective method for the treatment of tumor on the head of pancreas or ampulla. It’s a complex operation with multiple organs resected so that complications of anastomosis (such as pancreatic fistula, bile fistula, and anastomotic hemorrhage et al.) can not be avoided. Pancreatic fistula is the most important factor of them to influence the outcomes of operations. Early diagnosis and drainage could promote healing of pancreatic fistula, shorten the hospital stay time and reduce hospitalization expenses. Thus, Prophylactic abdominal drainage is considered as an effective method to decrease the morbidity and mortality of postoperative complications after pancreaticoduodenectomy. The drainage devices inserted by surgeons are different after pancreatic surgery, and their effects are unclear. Although drainage devices with different negative pressure are inserted after pancreaticoduodenectomy, it’s not clear whether the different mode of negative pressure drainage is the risk factors of postoperative complications. So, we conduct a test in vitro to analyze the character of two devise with different negative pressure drainage, and a RCT to exlpore the effect of negative pressure drainage with different modes on postoperative complications after pancreaticoduodenectomy, and the possible risk factors of pancreatic fistula after pancreaticoduodenectomy.Methods:1. Test of negative pressure in vitro:measure the negative pressure of Jackson-Pratt drains and Low Pressure Vacuum Drainage Systems with different volume of Saline.2. RCT:the patients were enrolled who should receive pancreaticoduodenectomy from April 15,2013 to January 31,2014 in the department of Pancreatic Surgery of Huashan Hospital Affiliated to Fudan University. The patients were excluded which complied with the following inclusion criteria:(1)patients rejected to participate in the clinical trial; (2)patients had contraindications of pancreaticoduodenectomy, such as coagulation abnormalities; (3)patients had participated in other clinical trials in recent 3 months; (4) tumor could not be resected after intraoperative exploration or pancreaticoduodenectomy was not appropriate for patients. Patients were randomized into two groups with different drainage devices before pancreaticoduodenectomy. Pancreaticoduodenectomy was performed with the mode of classic Whipple procedure or PPPD in this study. Single-layer end-to-side with external stent drainage was used in pancreaticojejunostomy. The texture of pancreas and size of the pancreatic duct were recorded. Two drainage tubes were placed in the superior and inferior of the pancreatic anastomosis respectively, and the drainage tube placed in the superior of the pancreatic anastomosis also went through the posterior of hepaticojejunostomy. The volume, color of drainage liquid, and the incidence of complications were recorded. Then, we explored the relationship between different drainage device and complications after pancreaticoduodenectomy, analyzed the risk factors of pancreatic fistula, and evaluated the value of different drainage mode. The definitions of complications are referred as the criteria of ISGPF.Result:1. Test of negative pressure in vitro:the negative pressure of Jackson-Pratt drains was 5.55±0.6kpa with Oml volume of saline, while the negative pressure of Low Pressure Vacuum Drainage Systems is 5.65±0.14kpa. The pressure changed with the volume of saline injected in both drainage devices, but the trends of change was statistically different (P<0.05). The negative pressure of Low Pressure Vacuum Drainage Systems changed less with 90ml volume saline below, compared with negative pressure of Jackson-Pratt drains. The negative pressure of Low Pressure Vacuum Drainage Systems was larger and relatively steadier than that of Jackson-Pratt drains.2. There were 66 patients included in this study. They were randomized into two groups:33patients in the Jackson-Pratt drains group, and the others in Low Pressure Vacuum Drainage Systems group. There was no significant difference of demographic data, texture of pancreas, pancreatic duct diameter or other clincopathological factors between two groups except drainage devices after surgery.3. Postoperative complications occurred in all patients, but most of them were grade Ⅰ complications without management, including Grade A pancreatic fistula, abdominal collection, pleural effusion and abdominal infection around drainage tubes. There were 18 patients with Grade Ⅱ complications in Jackson-Pratt drains group and 21 patients with Grade Ⅱ complications in Low Pressure Vacuum Drainage Systems group. Only 1 patients had Grade Ⅲ complications in all patients and he was in Low Pressure Vacuum Drainage Systems group. The patient had intestinal obstruction and received an operation. Nineteen patients had pancreatic fistula. Grade A pancreatic fistula occurred in 9 patients, and 10 patients had grade B pancreatic fistula. No patients had Grade C pancreatic fistula. Eight patients with Jackson-Pratt drains developed pancreatic fistula which accounted for 24% of all patients. Three of them had Grade A pancreatic fistula, and 5 patients had B pancreatic fistula. Eleven patients with Low Pressure Vacuum Drainage Systems developed pancreatic fistula,6 of them had grade A pancreatic fistula, and 5 patients had B pancreatic fistula. There was no significant difference in the incidences of pancreatic fistula between two groups. Tweenty nine patients had delayed gastric emptying (16 patients in the Jackson-Pratt drains group and 13 patients in the Low Pressure Vacuum Drainage Systems group. Five patients had peripancreatic collections. Two of them were in Jackson-Pratt drains group, and 3 patients were in Low Pressure Vacuum Drainage Systems group. There was not significant differences between two groups (P=1.000). Two groups of patients had no significant difference in the incidence of abdominal infection,15% in Jackson-Pratt drains group, and 18% in Low Pressure Vacuum Drainage Systems group respectively. Abdominal infection was found mainly around drainage tubes in this study, without obvious abdominal liquid infection. No postoperative hemorrhage or death was found.4. Drainage volume in Low Pressure Vacuum Drainage Systems group was significantly larger than that in Jackson-Pratt drains group in 2nd day and 4th day after surgery. However, there was no significant difference in drainage volumes between two groups in other days after PD. There was no significant difference in time to remove drainage, hospital stay days or readmission rate.5. Age (>60 years), operation time (>420min), diameter of pancreatic duct (≤ 3mm), soft pancreatic texture and PPPD operation were the risk factors of pancreatic fistula in the univariable analysis, but in the multivariate analysis, diameter of pancreatic duct (≤3mm) and the pancreatic texture were the independent predictive factor of pancreatic fistula.Conclusion:1. Most of postoperative complications are grade Ⅰ complications in this study, included pancreatic fistula, DGE, abdominal collection, pleural effusion, infection around drainage tubes and so on. There is no significant difference between the incidences of complications in patients with different negative drainage after pancreaticoduodenectomy.2. Low Pressure Vacuum Drainage Systems own higher and steadier negative pressure in vitro test.3. In the multivariate analysis, diameter of pancreatic duct (≤3mm) and soft pancreatic texture was the independent risk factor of pancreatic fistula after pancreaticoduodenectomy with pancreatic external stent.